(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Yale psychological studies"



LIBRARY 

Walter E. Fernald 
State School 




Waverley, Massachusetts 

No . ^rlA-S 



Vol. XXII PSYCHOLOGICAL REVIEW PUBLICATIONS Wh ?!f A No * 

No. 3 100 



Psychological Monographs 

EDITED BY 
JAMES ROWLAND ANGELL, University of Chicago 

HOWARD C. WARREN, Princeton University {Review) 
JOHN B. WATSON, Johns Hopkins University (/. of Exp. Psych.) 

SHEPHERD I. FRANZ, Govt. Hosp. for Insane (Bulletin) and 
MADISON BENTLEY, University of Illinois (Index) 



Yale Psychological Studies 

NEW SERIES— VOLUME II, No. 2 



Edited 
By 

ROSWELL P. ANGIER 

Professor of Psychology and Director of the 
Psychological Laboratory, Yale University 



PSYCHOLOGICAL REVIEW COMPANY 

PRINCETON, N. J. 
and LANCASTER, PA. 

Aghnt*: G. E. STECHERT & CO., London (2 Star Yard, Carey St., W. C); 
I.mpzig (Koenigstr.,37); Paris (16, rue de Conde) 




FRiNCLTON 

k UNlVLR.Sin> 

PRE.SS. 



TABLE OF CONTENTS 

Effects of Practice in the Discrimination and Sing- 
ing of Tones. By Edward Herbert Cameron, Ph.D., 
Assistant Professor of Education, Yale University. . . 159 

An Experimental Study of the Conscious Attitudes 
of Certainty and Uncertainty. By John Trum- 
bull Metcalf, Ph.D., Instructor in Psychology, Smith 
College 181 

Complex Reactions of the Dog: A Preliminary 
Study. By Arthur Howard Sutherland, Ph.D., In- 
structor in Psychology, Yale University 241 

An Experimental Study of Mental Capacities of . ^> * 
School Children, Correlated with Social ^ 
Status. By Horace Bidwell English, Ph.D., In- 
structor in Psychology, Wellesley College 266 



EFFECTS OF PRACTICE IN THE DISCRIMINATION 

AND SINGING OF TONES 

Edward Herbert Cameron, Ph.D. 

Assistant Professor of Education 

Yale University 

A study (i) of the effects of practice on singing tones at two different 
pitch levels near the upper and lower limits of the range of the male voice; 
and (2) of the relation of these activities t(o discrimination of tones. 

Practice in singing a tone of a certain pitch level resulted in improvement 
in accuracy of pitch of the tone in the cases of four out of six subjects. 
This improvement was not present in the singing of tones at a different level 
from that in which practice had taken place. Practice in singing tones of a 
certain pitch was followed by improvement in discrimination of tones of the 
same pitch but no improvement in discrimination took place at the other 
pitch level. Improvement in the uniformity of the practised sung tone was 
transferred to the other level. 

Perhaps in no other sphere is the usual voluntary response to 
a sensory stimulus more precise and unvarying than in the repro- 
duction of tones by singing. The vocal production of tones is, 
therefore, peculiarly adapted for the study of sensori-motor 
habits. While the sensory processes involved in such habits have 
been studied in detail, relatively little investigation has been made 
of the related motor processes. In a former paper the writer has 
shown some of the characteristics of sung tones and their changes 
under varying conditions of attention. Berlage (1) and Miles 
(7) have confirmed this work in some respects and expanded it 
in others. 

The object of the investigation now presented was fourfold : 

( 1 ) To determine the relationship between ability to discrimi- 
nate tones of a certain pitch and the ability to sing tones of the 
same pitch. 

(2) To determine the changes which take place in the singing 
of tones of a certain pitch during the course of a long practice 
series. 

(3) To determine what influence, if any, practice in singing 



160 EDWARD HERBERT CAMERON 

of a certain pitch has on the ability to discriminate tones (a) of 
the same pitch and (b) of a different pitch. 

(4) To determine what effect, if any, practice in singing tones 
of a certain pitch has on singing tones of a different pitch. 

For carrying out such an investigation it is necessary to have 
(a) a means of testing sensory discrimination of tones of dif- 
ferent pitch; (b) a means of recording the pitch of a sung tone. 

The difficulties of a method for the former purpose are well- 
known. Seashore's report (10) recommends the use of standard 
tuning forks of slightly different pitches and appropriate resona- 
tors. This method was not used in the present instance because 
of difficulty in obtaining forks of the range of the human (male) 
voice, and because of certain advantages of manipulation to be 
gained by the method actually adopted. 

In the method used the source of the tones, both for purposes 
of testing discrimination and for reproduction by singing, was 
electric tuning forks. Four such forks, two of approximately 
100 v. d. and two of approximately 225 v. d. frequency were 
used. These forks were placed in circuit with a single telephone 
receiver of the watch case variety, which was placed in a room 
adjoining that in which the forks were placed. The prongs of 
one of each pair of forks were fitted with adjustable weights by 
means of which the pitch was varied above or below that of the 
other fork of the same range, the pitch of the latter being kept 
constant. The variations in pitch were measured indirectly by 
noting the position of the weights on a millimeter scale attached 
to the base at one side of the fork. To facilitate the reading of 
the position of the weights one of them was provided with a 
pointer which projected to a point just above the scale. The 
pitch for the various readings on the scale was ascertained by 
placing the forks in circuit with an electric marker recording on 
a smoked paper belt. 

The telephone and forks were connected by means of a Pohl 
commutator, the cross bars at the base having been removed. 
The semi-circular rods were kept out of the mercury cups when 
at rest by a pair of springs attached to the base. When the 
switch was in this position the circuit to the telephone was inter- 



DISCRIMINATION AND SINGING OF TONES 161 

rupted. The current passed through the telephone and the 
standard fork when the free ends of the rods of the commutator 
were pressed in one direction, and through the telephone and 
variable fork when pressed in the opposite direction. 

In the earlier investigation mentioned above there was used 
for the recording of sung tones a telephone receiver with mica 
diaphragm to which were attached magnifying levers. This 
method was discarded in the present investigation. A graphic 
record of the sung tone was obtained through the use of an ordi- 
nary voice key, similar to the Cattell model, but without any 
electro-magnetic attachment. The diaphragm is of very thin 
mica and is readily thrown into vibration by a tone sung into 
the air chamber. The voice key was placed in circuit with an 
electric marker of the Deprez pattern. 

The sung tones were recorded on a long smoked paper belt 
running on two drums which were turned by hand. The time 
was recorded by an electric marker in circuit with a Kronecker 
interrupter regulated to intervals of ioo sigmas duration. 

Before beginning the experiment the following preliminary 
tests were made: 

(i) A test of the capacity of the telephone to register cor- 
rectly the tone of the fork and respond to slight variations of 
tone. For this purpose recording levers were attached to the 
diaphragm of the telephone and the record thus obtained was 
compared with one obtained simultaneously from the fork 
through an electro-magnetic marker. There was no discoverable 
difference between the vibration rate of the fork and that of the 
telephone when recorded in this way. 

(2) A similar test was made of the apparatus for recording 
the sung tones. A mica disk was attached to one of the prongs 
of the fork and placed in front of the air chamber of the voice 
key. In this manner records could be obtained of the vibration 
rate of the fork when transmitted through the voice key and 
comparison of these records with the records obtained at the 
same time through an electric marker showed no appreciable 
difference. 

The tones obtained from the forks through the medium of the 



162 EDWARD HERBERT CAMERON 

telephone were not pure tones and differed somewhat in quality, 
especially in the case of the pair of forks of lower pitch. The 
intensities of the tones of each pair of forks were apparently the 
same. The qualitative differences did not remain constant, but 
varied with differences of adjustment of the forks, especially 
differences in the distance of the adjustable button from the 
spring of the fork. In order to ensure the best possible adjust- 
ment, the experimenter made use of a second telephone which 
could be connected with the forks by means of a switch. In this 
way he could test the tones at will and introduce such variations 
in adjustment as to secure better tones, when necessary. 

At best, however, the qualitative differences were nearly always 
sufficient to enable the observer to distinguish the two tones. 
Great precautions had, therefore, to be taken in the discrimination 
tests to make sure that the results were due to differences in pitch 
and not differences in quality. Furthermore, on these and other 
grounds, all claims that the results give an absolute pitch limen 
for the subjects tested must be abandoned. It was sought only 
to obtain a limen which should hold under the conditions of this 
experiment- and which should serve as a basis of comparison be- 
tween the various observers who were tested. 

The subjects who took part in the investigation were as fol- 
lows : Professor Angier and Dr. Frost of the Yale Psychological 
Laboratory; Dr. Metcalf, laboratory assistant; and Messrs. Reed, 
Molby and Avey, graduate students in psychology. These sub- 
jects will be referred to hereafter as A., F., M., R., My., and Ay. 
respectively. All of these subjects except A. are of average or 
slightly more than average musical ability. M. and R. have had 
some training in the use of musical instruments, while My. and 
Ay. have been trained in singing to a slightly greater degree than 
the average person. A. and F. have had no musical training 
whatever. F. sings only infrequently. A. belongs to that class 
of individuals who cannot "carry a tune" correctly, though his 
inability in this respect is not as marked as in many cases. 

All six subjects were given a preliminary test (i) of discrimi- 
nation of pitch at both the higher and lower levels; (2) of ability 
to reproduce, by singing, tones of both the higher and lower 



DISCRIMINATION AND SINGING OF TONES 163 

levels. Three of the subjects, A., F., and Ay., were then prac- 
tised for a long time in singing the higher tone; the other three 
subjects, M., R., and My., were practised in the singing of the 
lower tone. After the practice in singing had stopped all sub- 
jects were re-tested as in the preliminary stage of the experiment. 

In determining the capacity for pitch discrimination the fol- 
lowing procedure was adopted. The observer sat in a partially 
darkened room with the telephone pressed lightly to the ear. A 
ready signal was given by means of a momentary flash of light 
from an electric lamp which was placed in front of the subject, 
but at some ten feet distance. A moment later the standard tone 
and the comparison tones were given for periods of two seconds 
each with a two second interval between them. The observer 
made known his judgment by pressing a key which was connected 
with a sound hammer in the experimenter's room. Judgments 
w T ere classified as (1) same or doubtful; (2) higher; or (3) 
lower. 

The pitch of the variable fork was varied by gradual steps 
below that of the standard fork and the threshold was regarded 
as having been reached at the point where the observer gave five 
correct judgments in succession. To avoid as far as possible 
unintentional judgments on the basis of the differences in quality 
of the two tones the standard tone was frequently given after 
the comparison tone. Furthermore, when it appeared that the 
threshold had been reached, the pitch of the variable fork was 
occasionally made higher instead of lower than that of the stand- 
ard. In this way the experimenter could test whether the ob- 
server was meeting the requirements of the experiment by relying 
solely on differences of pitch in making his judgments. If it was 
found that this was not the case, as happened in several instances, 
the observer was cautioned and usually adopted a more critical 
attitude toward his judgments. Before recording any results in 
these experiments two periods of about twenty minutes each were 
allowed each observer for adaptation to the conditions of the 
experiment. The results of the preliminary test of discrimination 
of pitch follow (Table I). 

The table gives the discrimination thresholds in vibrations per 



i6 4 



UPWARD HERBERT CAMERON 



second at both lower and higher pitches for each of the six sub- 
jects. Five trials on different days were made for each subject. 
In most cases the subjects expressed themselves as having no 
difficulty in distinguishing between the pitch and qualitative dif- 
ferences of the tones. In some instances, however, there was so 
much confusion that the test for that day was abandoned. While 
the confusion manifested may have been aggravated by the quali- 
tative differences in the forks, it cannot be wholly charged to 
this cause, since other investigators have found similar results 
under more ideally perfect conditions. 

Table I 
Discrimination — Threshold before practice. 

Low (S = 100.3 V.D.) High (S = 225.1 V.D.) 





A. 


F. 


Ay. 


M. 


R. 


My. 


A. 


F. 


Ay. 


M. 


R. 


My. 


I. 


1.9 


1.9 


1.9 


1-3 


1.3 


1.9 


3-2 


2.6 


2.6 


2.0 


2.6 


2.6 


2. 


2.4 


2.4 


1.9 


1.9 


1-3 


1.9 


3-2 


2.6 


2.0 


2.0 


2.6 


2.0 


3- 


2.4 


1-9 


2.4 


1.3 


1-3 


1.9 


3-8 


2.6 


2.0 


2.0 


2.6 


2.0 


4- 


2.4 


2.4 


2.4 


1-3 


1-3 


1.3 


3-8 


2.6 


2.0 


1-5 


2.6 


2.0 


5- 


1.9 


2.4 


2.4 


1.9 


13 


1.9 


3-2 


2.6 


2.0 


1-5 


2.6 


2.0 


Av. 


2.2 


2.2 


2.2 


1.4 


1-3 


1.8 


34 


2.6 


2.1 


1.8 


2.6 


2.1 


A.D. 


0.2 


0.2 


0.2 


0.3 


0.0 


0.2 


0.3' 


0.0 


0.2 


0.2 


0.0 


0.2 



It will be seen from the table that the thresholds are greater 
for the higher than for the lower level. The average for all sub- 
jects at the lower level is 1.9 and for the higher level 2.4. These 
results agree with those of other investigators who have found 
that the threshold for pitch discrimination at various ranges does 
not increase proportionally to the vibration rate of the tones. 

It will be noted that A., who is the decidedly unmusical sub- 
ject, gives a record for pitch discrimination which does not differ 
largely from the records of the other subjects. At the lower 
level his average discrimination is the same as that of the other 
subjects. A.'s discrimination at the higher level is the highest 
of the six records, but it is not greatly different from that of two 
of the others. 

In carrying out the experiments in singing the subject w r as 
given a ready signal (light) and a moment later the tone of the 
standard fork was sounded for two seconds. The subject was 
instructed to wait a moment after the tone stopped sounding and 
then sing into the mouthpiece a tone of the same pitch and ap- 



DISCRIMINATION AND SINGING OF TONES 165 

proximately the same duration. Berlage and Miles have found 
that the pitch of a sung tone is somewhat modified by the vowel 
sound used in singing. The singer was, therefore, instructed to 
modulate his voice to the vowel sound "a" as in "ah." After 
allowing a few trials for the sake of adapting the subject to the 
condition of the experiment, five records were obtained from 
each of the subjects for each of the two tones. The records were 
read by means of the apparatus described in the former paper 
and in a similar manner to that of the earlier investigation. The 
results, expressed in vibration rate per second for each succeed- 
ing period of a tenth of a second's duration, are given in Table 
II. Figures are given for only the first second during which the 
tone was sung as this period is fairly representative of the whole 
tone. 

In the column marked pitch is given the vibration frequency 
per second of each tenth of second for a continuous period of 
one second from the beginning of the tone. The figures given 
represent, therefore, the actual number of vibrations on the 
record multiplied by ten, so as to express the pitch for each period 
of 100 sigmas in the conventional unit of number of vibrations 
per second. Since the first two or three-tenths of second of the 
tone are marked periods of adjustment they do not represent the 
main tone correctly. Two averages are, therefore, given; the 
first for the earlier two or three-tenths of second, and the second 
for the remainder of the tone. An examination of the figures for 
the earlier part of the tone shows that this average is usually, 
though not invariably, lower than that for the rest of the tone. 
Usually the singer begins much lower than the final average and 
in two or three-tenths of second reaches a pitch much higher 
than the one to which he finally settles down. This process of 
sliding up to the tone is much less marked in singing the lower 
tone and at times it seems as if the process were reversed in 
these tones, the singer beginning too high and sliding down to 
the pitch finally adopted. The upper of the two figures in the 
third column for each tone gives the extreme variation of these 
earlier tones (the difference between the highest and lowest 
pitch). Besides this lack of uniformity in the earlier part of the 



1 66 



EDWARD HERBERT CAMERON 



tone there are other though usually less marked variations in 
the pitch from any given one-tenth of a second to the next 
throughout the rest of the tone. The lower of the two figures 
in the third column for each tone gives the average deviation for 
the latter part of the tone and is, therefore, a measure of the 
degree of lack of uniformity of pitch in the main tone. 

Turning from these general features of Table II to the indi- 
vidual results, it will be seen that all the subjects show much lack 
of uniformity both in beginning the tones and in continuing 
them. Roughly, the amount of this lack of uniformity is pro- 
portional to the vibration frequency ; the variations in the higher 
tones being much greater than in the lower. Subjects A., F., 
and My. are not even approximately correct in singing the low 
tone, but in all other cases there is a fair degree of approximation 
to the standard. In these cases it seems clear that the difficulty 
is to be accounted for in large measure by the inability of the 



Subject 



A. 



Table II 
Singing before practice. 













Low Tone (S = 


= 100.3 


V.DJ 












I 


II 


III 


DevT 


IV 




V 




Pitch Av. 


Dev. 


Pitch Av. 


Dev. 


Pitch 


Av. 


Pitch Av. 


Dev. 


Pitch | 


Av. | 


Dev. 


140 | 




140 I 




133 






119 






127 






140 




149 


144.5 


9 


149 


141 


16 


141 


130 1 


22 


129 






149 1 143 


9 


143 






147 






142 






131 


129.0 


4 


153 






149 






152 






156 






128 






146 






148 






147 






149 






126 






144 






148 






148 






151 






127 






148 






148 






149 






150 






129 






146 






146 






146 






150 






127 






146 






148 






148 






151 






127 






148 


147 

1 


2.0 


145 
1 


146.8 


1.7 

High 


149 
Tone 


148.3 
(S = 


1-3 
= 225.1 


155 
1 V.D. 

j 


149-3 
) 


2.4 


129 


127.6 


1.2 


188 






180 






185 






175 






172 






202 






215 






211 






224 


199.5 


49. 


215 


47 


18.2 


210 


200.0 


32 


218 


204.3 


38 


220 


253.3 


35 


224 






219 






213 


'- 




216 






218 






220 






216 






217 






224 






215 






220 






219 






216 






220 






220 






220 






216 






215 






216 






215 






220 






215 






219 






215 






220 






218 






217 






217 






21.S 






215 






218 






218 






217 


217.7 


1.8 


217 


217.6 


2.5 


215 
218 
218 


217 


1.8 


215 


219.4 


1.8 


215 


216.6 


1.2 



DISCRIMINATION AND SINGING OF TONES 



167 



Subject F. 













Low Tone (S = 100.3 V.D.) 












I 




II 


III 


IV 




V 




Pitch 


Av. 


Dev. 


Pitch 


Ay. 


Dev. 


Pitch 


Av. 


Dev. 


Pitch 


Av. 


Dev. 


Pitch 


Av. 


Dev. 


143 






134 






121 






150 






125 






150 


146.5 


7 


143 






139 






140 






133 






140 






150 


142.3 


16 


148 


136 


27 


146 


145-3 


4 


146 


134-7 


21 


144 






151 






144 






146 






141 






141 






152 






146 






143 






142 






143 






143 






144 






146 






143 






141 






144 






142 






143 






146 






140 






143 






145 






143 






145 






145 






150 






141 






147 






145 


/ 




143 


142. 1 


1.8 


145 


146.9 


3-6 
High 


142 
Tone 


144-9 
(S = 


1.8 
= 225.1 


143 
V.D. 


144-4 
) 


1.9 


143 


143-6 


1-5 


18/ 






152 






168 






171 






190 






213 






178 






201 






209 






216 


( 




227 


209 


40 


197 


175.6 


45 


219 


196.0 


5i 


215 


198.3 


44 


220 


208.3 


30 


216 






220 






220 






212 






214 






220 






222 






218 






214 






226 






217 






220 






218 






218 






216 






219 






217 






220 






218 






218 






220 






223 






217 






220 






220 






220 






215 






219 






217 






220 






219 


218.7 


1.5 


221 


221.9 


3.0 


220 


218.9 


0.9 


219 


216.8 


2.2 


220 


219. 1 


2.7 



Subject Ay. 













Low Tone ( 


:s = ] 


[00.3 V.D.) 












I 


II 


III 


IV 


V 


Pitch 


Av. 


Dev. 


Pitch 


Av. 


Dev. 


Pitch 


Av. 


Dev. 


Pitch 


Av. 


Dev. 


Pitch 


Av. 


Dev. 


112 






105 




1 


106 






116 




1 


no 






109 






103 


104 


! 2 


107 


106.5 


1 


107 


111.5 


9 


107 


108.5 


7 


90 


105-7 


16 


109 




1 


no 






103 






113 






117 






108 






109 






105 






108 






105 






107 






104 






107 






in 






105 






108 






106 






109 






no 






102 






106 






109 






103 






104 






102 






104 






106 






108 






109 






105 






106 






109 






107 






106 






108 


106.3 


3-6 


108 


107.0 


1.3 

High 


108 

1 
Tom 


107.6 
1 

! (S = 


1.8 
= 225.1 


107 
V.D. 


106.1 

1 
) 


1.9 


109 


108.8 


2.1 


179 






221 






219 






236 






222 






231 






230 


225.5 


9 


228 


223.5 


9 


228 






23'4 


228.0 


12 


219 
218 


209.7 


40 


211 






216 






217 


223.7 


19 


220 










221 






227 






221 






228 


• 




223 






221 






225 






223 






221 






226 






221 






228 






219 






221 






219 






221 






223 






226 






230 






224 






230 






227 






219 






225 






221 






223 






224 






223 






229 






224 


222.1 


24 


213 


221.0 


3-6 


222 


224.O 


2.6 


2l8 


223.7 


2.3 


225 


224.9 | 
I 


3-1 



IU V 



EDWARD HERBERT CAMERON 



Subject M. 



Low Tone (S = 100.3 V.D.) 





I 




II 


III 


IV 


V 


Pitch 


Av. 


Dev. 


Pitch 
98" 


Av. 


Dev. 


Pitchl 


Av. 


Dev. 


Pitch 


Av. 


Dev. 


Pitchl 


Av. | Dev. 


IOI 






85 






85 






109 


1 


99 






103 


100.5 1 


5 


90 






IOI 


93 


16 


112 


1 10.5 1 3 


98 


99-3 


3 


103 






100 


91.6 


15 


IOI 






108 






95 






106 






IOI 






IOO 






108 






IOI 






102 






95 






99 






103 






100 






103 






100 






97 






108 






100 






IOI 






99 






100 






108 






103 






100 






99 






99 






102 






103 






102 






IOI 






IOI 






109 






103 


102.1 


2.2 


105 

1 


102.8 


1.6 
Higl 


102 
1 Ton* 


99.6 

2 (S = 


1.6 
= 225.1 


102 
V.D. 


99-9 
) 


1.2 


109 


106.9 


2.1 


230 






221 






201 






218 






225 






222 


226 


8 


227 






226 






220 


219 


2 


220 


222.5 


5 


225 






233 


227 


12 


228 


218.3 


27 


220 






220 






221 






228 






229 






222 






222 






225 






224 






221 






219 






219 






220 






225 






221 






223 






223 






228 






220 






222 






221 






221 






225 






224 






221 






224 






224 






222 






221 






224 






220 






220 






226 


224 


2.3 


228 


224.3 


2.3 


220 


222.6 


2.3 


224 


221.6 


1.6 


224 


221.6 


1.6 



Subject R. 













Low Tone 1 


:s = ■ 


[00.3 V.D.) 












I 


II 


III 


IV 


V 


Pitch Av. | 


Dev. 


Pitch 


Av. 


Dev. 


Pitchl Av. 


Dev. 


Pitch Av. 


Dev. 


Pitchl 


Av. 


Dev. 


105 1 






102 






107 | 


1 




105 






IOI 






no 






102 






99 






107 






102 






115 


IIO.O 


10 


105 


103.0 


3' 


100 




8 


109 


107.0 


4 


102 


101.7 


1 


112 






107 






105 






no 






103 






Il6 






106 






107 






no 






103 






112 






106 






107 






no 






IOI 






115 






107 






105 






109 






103 






Il6 






104 






107 






112 






IOI 






115 






102 






108 






112 






104 






112 


1 14.0 


1.7 


105 


105.4 


1.4 
High 


1 09 
Tone 


106.0 

1 

(S = 


2.0 

1 

= 225.1 


no 

v.d.; 


110.5 
) 


1.6 


IOO 


102. 1 


1-3 


177 






215 






220 






186 






220 






190 






229 


222.0 


14 


229 






210 






232 


228 


12 


212 


193-0 


1 35 


224 






235 


228.0 


15 


218 


204.6 


32 


230 






2l6 






234 






235 






222 






235 






224 






231 






231 






229 






230 






2l8 






235 






236 






226 






232 






220 






230 






239 






230 


I 




228 






225 






238 






237 > 






228 






23'5 


■ 




217 






234 






23'8 






230 




■ 


| 232 






220 


220 


1 3-0 


238 


233-0 


3-5 


240 


236.6 


2.1 


233 


228.3 


2-5 


j 230 


231.5 


2.0 



DISCRIMINATION AND SINGING OF TONES 



169 



Subji 


>ct Mv 




































Low T 


one (S = 100.3 V.D.) 












I 


,| 


II 




Ill 




IV 




Pitch | 
119 


V 




Pitch 


Av. Dev. 


Pitch 


Ay. 


Dev. 


Pitch 


Av. 


Dev. 


Pitch 


Av. 


_Pev ! _ 


Av. | 


Dev. 


121 






110 






118 






113 








120 


120.5 


1 


112 




1 


119 






114 






116 






121 






114 


1 1 2.0 


4 


120 


1 19.0 


2 


118 


115.0 


5 


119 


II7.5 


3 


122 






114 




1 


122 






119 






118 






121 






118 






122 






120 






121 






120 






118 






122 




i 


119 






118 






1 22 






119 






121 






119 






120 






120 






120 






120 




1 


120 






120 






120 






119 






122 






120 






121 






123 


121.2 


0.9 


120 

I 
I 


1 18.3 

1 

1 


1.4 
High 


120 
Tone 


121.3 

(S = 


0.8 
= 225.1 


120 
V.D. 


1 19.6 
) 


0.5 


120 

j 


1 19.6 


1.0 


215 






215 






214 






209 






221 






225 






225 






217 






215 






224 


227.5 


3 


226 


220.0 


10 


225 


220.0 


10 


22A 


218.3 


10 


220 


214.7 


11 


222 






226 






226 






221 






217 






225 






225 






226 






223 






221 






227 






224 






226 






221 






221 






223 






228 






218 






221 






219 






22Z 






226 






224 






223 






224 






226 






224 






222 






221 






219 






222 






220 


224.9 


i-9 


225 


224.0 


1.8 


222 

1 


221.7 


0.8 


221 


219,7 


1.8 

1 

! 


225 


224.2 


1.6 



singer to distinguish the fundamental tone. Indeed, it would 
appear from the results that A. and F. are perhaps both respond- 
ing to a prominent overtone. 

In carrying out the practice series of experiments subjects A., 
F., and Ay. were practiced in the singing of the low tone. These 
practice experiments consisted in the singing of the standard in 
the manner already described twenty times daily for a period 
extending over several months. In this way a total number of 
approximately one thousand tones were sung by each subject. 
M., R., and My. were practiced in a similar manner with the 
high tone. When the practice period was about half finished the 
method was somewhat modified. The subjects were now asked 
to sing the tone while the standard tone was still sounding. At 
this point in the experiment and, indeed, from near the beginning 
of the practice series, all the subjects, except A., were approxi- 
mating the standards. Singing the tone in unison with the stand- 
ard resulted in some improvement with the other subjects, but 
A's. record continued to be very irregular and inaccurate. 



I/O 



EDWARD HERBERT CAMERON 



In A's. case, therefore, a more radical attempt was made to 
obtain exactness of reproduction of tone. The tone of the stand- 
ard fork was now made by striking the fork with a rubber ham- 
mer and reinforced by an appropriate resonator. Under these 
conditions A. responded with tones which were more nearly the 
tone of the standard, but very inconsistently and with no close 
approximation of the standard tone, although persistent attempts 
to obtain correct results were made. 

Before obtaining the final results of singing the standard tones 
after practice, one hundred practice trials were made in the man- 
ner originally adopted — that is, with a pause between the stand- 
ard tone and the beginning of the singing. The results of the 
final series of five tones at both levels for each of the subjects 
are shown in Table III. 

For more convenient comparison, the results of Tables II and 
III are condensed in Table IV. Columns V! represent the aver- 
age deviation of the sung tones from the standard before and 



Subject A. 



Table III 

Singing after practice. 
Low Tone (5 — 100.3 V.D.) 





I 






II 






III 






IV 






V 




Pitch Av. 


Dev. 

1 


Pitch 
118 


_Av.__ 


_Dey,_ 


Pitch Av. 


Dev. 


Pitch Av. Dev. 1 Pitch Av. 


Dev. 


122 ! 


115 






118 






113 






120 


121.0 


1 2 


118 






118 






120 


1 1 9.0 


2 


116 


1 14.5 


3 


120 






113 


149.7 


5 


121 


1 18.0 


6 


119 






113 






118 






118 






117 






119 






117 






120 






120 






119 






120 






118 






120 






116 






120 






117 






113 






119 






117 






117 






118 






114 






116 






119 






121 






120 






113 






120 






118 






117 






119 






112 






118 


1 18.9 


1.2 


ii5 


117.6 


i.3 

Higl 


119 
1 Tom 


118.5 
1 (S = 


1.3 

= 225.1 


118 
V.D. 


118.7 
) 


0.9 


ii5 


1 14.4 


1-3 


153 






141 






148 






144 






143 






167 






166 






182 






169 






172 






175 


165.0 


22 


173 


160.0 


32 


183 


171.0 


35 


175 


162.7 


3i 


176 


163.7 


33 


170 






175 






179 






178 






165 






172 






178 






175- 






180 






170 






174 






181 






180 






177 






170 






172 






175 






178 






181 






172 






172 






176 






180 






178 






170 






172 






174 


• 




176 






180 






173 


^ 




173 


1 72. 1 


0.8 


178 


176.7 


2.0 


180 


178.3 1 


1-7 


176 

1 


178.6 


1-7 


174 


170.6 


2.1 



DISCRIMINATION AND SINGING OF TONES 



i/i 



pubjcct 1 






II 


Low Tone (S : 


= 100.3 


v.d.; 


) 










f i 




in 




IV 


v 


JPitchl Av. 


Dev. 


Pitch 


Av. 


Dev. 


Pitch 1 Av. 


Dev. ; 


Pitch 1 Av. 


Dev. i 


Pitch | Av. Dev. 


981 


i 99 






103 






104 






99 






IOO 




100 






103 






102 






101 






; ioj 


IOO.O 


4 


103 


100.7 


4 


102 


102.7 


1 


102 






103 


IOI.O 


4 


102 






102 






103 






101 






104 






102 






103 






103 






101 






102 






ior 






102 






103 






101 






102 












103 






103 






102 






102 






104 






103 






103 






102 






103 






104 






104 






104 






102 






102 






103 


102.6 


0.9 


102 


102.7 


0.6 

Hi g : 


104 
1 Ton 


103.3' 

e (S : 


0.4 

! 
= 225.1 


103 

V.D. 


102.0 

) 


0, 


103 


102.6 


0.7 


200 






192 






210 






207 






215 






212 






210 






218 






210 






220 


217.5 


50 


22 


213.0 


-7 


221 


210.7 


29 


220 


216.0 


10 


220 


212.3 


13 


220 






22 






221 






220 






222 






220 






221 




1 


221 






221 






220 






: 223 






221 






215 






221 






222 






222 






221 






223 






220 






222 






222 






223 






220 






221 






222 






\ 220 






222 






225 






222 






220 






221 






22 


221.7 


0.6 


222 


222.4 


1.6 


220 


220.7 


0.6 






! 


| 223 


221.4 


1.1 



Subject Ay 













Low Tone (S = 100.3 V.D.) 












I 




II 


Ill 






IV 






V 




3 itch 1 Av. 


Dev. 


Pitch 


Av. 


Dev. 


Pitch Av. 


Dev. 


Pitch 


Av. 


Dev. 


Pitchl Av. 


Dev. 


99 | 




102 






101 






08 






98 






100 


995 


1 


103 


102.5 


1 


100 


100.5 


1 


95 


06.5 


3 


99 


98.5 


1 


100 






102 






101 






100 






08 






99 






102 






08 






101 






98 






100 






IOO 






IOO 






101 






08 






100 






IOO 






IOO 






102 






08 






100 






98 






98 






102 






99 






101 






IOO 






101 






101 






99 






101 






IOO 






101 






99 






99 


08.8 


0-7 


100 


IOO. I 


04 


101 


100.4 


1.0 ; 

High 


IOO 

Tone 


999 

! (S = 


0.9 
: 225.1 


101 

V.D.) 


100.9 


0.7 ! 


101 






210 






191 






210 






217 






215 






216 






220 


215 


19 


220 


2150 


IO 


220 


218.5 


3 


219 


217 


4 


2V) 


215.0 


9 


222 






218 






220 






218 












220 






221 






220 






220 






220 






21 






220 






220 






220 






219 






222 






j 20 






220 






220 






217 






220 






220 


220.0 


O.I 


2iq 






220 






220 






221 












22] 






218 












221 






220 






22] 






222 






220 


220.0 


1.4 

1 


210 


220.0 


, ' : i 


22 r 
220 




• 


219 


220.0 
1 


0.1 

! 


219 


219.6 


1.0 



EDWARD HERBERT CAMERON 



Subject M. 












Low Tone (S = 100.3 V.D.) 












I 


II 


III 


IV 


V 


Pitch 


Av. 


Dev. 


Pitdh Av. 


Dev. 


Pitch] Av. 


Dev. 


Pitch | Av. Dev. 


Pitch 


Av. 


Dev. 


105 






107 | 




107 






105 






109 






106 


105.5 


1 


103 I 105,0 


4 


107 






107 


106.0 


2 


106 


107.5 


3 


104 






108 | 




108 


107.3 


1 


106 






108 






105 






109 1 




107 






107 






109 






102 






106 




106 






107 






107 






103 






108 | 




105 






108 


' 




109 






107 






107 I 




106 






107 






109 






102 






109 1 




107 






108 






108 






102 






105 ! 




105 






108 






109 






102 


1034 


1-5 


105 1 107.1 


14 


105 


105.9 


0.7 


106 


107.0 


0-7 


109 


108.5 


0.6 








1 

1 

1 


High 


l Tone 


: (S = 


= 225.1 


V.D.) 


I 










220 






215 




216 






219 






219 






223 






220 




225 






222, 






222 






225 


222.7 


5 


221 | 218.7 


6 


228 


223.0 


12 


225 


222.3 


6 


224 


221.7 


5 


228 






225 1 




228 






223 






222 






225 






225 1 




229 






225 






224 






228 






225 1 




230 






224 






224 






225' 






226 




229 






224 






224 






228 






225 




229 






226 






226 






228 






227 




230 






226 






223 






229 


227.3' 


13 


228 | 225.7 

1 


1.0 


228 


229.0 


O.I 


228 


225.0 


1-3 


225 


224.0 


O.I 





ci My 






Pito 


Ay. Dev.' 


i 4 i 






U3 


142 


j 


140 






1 50 






146 






"142 






143 






147 






143 




143 144.3 


2.1 


231 




239 


5-0 


-37 






23,2 






M 






2 20 






-4 






228 






-30 




>8 


229.9 


2.6 



DISCRIMINATION AND SINGING OF TONES 



Low Tone (S = 100.3 V.D.) 



173 



II 



Fitch I Av. Dev. 



141 
144 
139 
148 

139 
U3 
140 
146 
142 
142 



I 



228 
228 
225 
226 

231 
230 

22^ 
228 
224 

227 



142.5 



142.6 



227.0 



227.3 



2.8 



High 





III 




Pitdhl Av. 


Dev. 


137 






i45 


141 


8 


137 






142 






142 






143 






143 






142 






138 






145 


I4I.5 


2.0 



143-0 



IV 

Pitch I Av. 

"T48" 
U3 
138 
142 
146 
141 
146 

145 
142 

143 



Dev. 



143 6 



2.1 



1 Tone (S = 


= 225.1 


V.D.) 

1 


230 1 




229 ! 


222 




227 


228 | 226.0 


8 


224 


223 




227 


226 




221 


224 | 




230 1 


227 




230 | 


223 




223 


220 




229 


229 | 225.1 


2.4 


222 



226.7 



226.0 



10 



1.8 



34 



V 



140 
138 

145 
148 
142 

145 
MS 
140 

143 
142 



218 
213 
220 

213 
222 
211 
221 
216 
218 
212 



139-0 



143-8 



215.5 



216.6 



Pitch Av. I Etev. 



2.1 



3-6 



Table IV 



Subj 






Low 




High 




A 


Before 
After 


v, 

43-5 
17-3 


v 2 

1-7 
1.2 


v 3 

1.2 

3-6 


V, 

7-4 
49-8 


Vo 

1.8 

i-7 


v 3 

40.2 
30.6 


•F 


Before 
After 


44.0 
2.3 


2.1 
0.6 


15 
3 


6.0 
3-7 


2.1 
1.0 


42.0 
17.0 


Ay 


Before 
After 


6.9 
0.6 


2.1 
0.8 


7 
1.4 


2.0 

5-2 


2.8 
0.8 


17.8 
9.0 


M 


Before 
After 


2.4 
i-7 


1.8 
1.2 


8.4 

4-4 


2.3 
1-3 


2.2 
0.9 


10.8 
10.2 


R 


Before 
After 


7-3 
6.1 


1.6 
1.0 


5-2 

2.2 


6.8 
1.6 


2.6 
0.8 


21.6 
6.8 


My 


Before 
After 


19.7 


0.9 

2.2 


3-0 
5.0 


2.2 
3-3 


1.6 
2.8 


8.8 
5-8 



after practice for both high and low tones. In arriving at these 
averages the earlier two or three-tenths of a second of the tone 
were not taken into account since, as we have already seen, they 
are not representative of the tone as a whole, but of a period of 
initial adjustment. Columns V 2 give the average of the mean 
variation- of the five tones for each subject before and after 
practice. These figures, therefore, represent the degree of the 



174 EDWARD HERBERT CAMERON 

steadiness of the tone as a whole, disregarding as before the 
initial period. In columns V 3 are arranged the average of the 
variations at this initial period for the five high and five low 
tones before and after practice. The figures of V 1 show that A., 
F., and Ay., who practiced with the low tone, reduced the error 
in singing at that level. In the case of A., however, the improve- 
ment is more apparent than real, for at no time does he approxi- 
mate the standard. His records from day to day show no gradual 
improvement as in the case of the other subjects and no constancy 
of results. The introspective records show that this subject 
toward the end of the practice series responded in a hit-or-miss 
fashion with a tone near the lower level of his voice register. 
When this subject began his practice on any given day with a 
sung tone of a certain pitch, his subsequent tones were usually 
of about the same pitch. This tendency to persist in repeating 
a tone of the same pitch when once begun was found to be char- 
acteristic of non-musical subjects in my former study of singing 
reactions (3) and also in the work of Miles (7). It would ap- 
pear that such persons in the absence of complete organization 
of the auditory and vocal motor factors rely mainly upon a 
memory of the kinaesthetic sensations as cue to further repro- 
ductions of the tone. 

In the case of F. who began with an error as great as that of 
A. and in the same direction, a tone was sung on the second day 
of practice which approximated the standard and improvement 
thereafter was of a steady character, quite in contrast with the 
erratic nature of A.'s record. Ay. began with a tone approxi- 
mating the standard and made steady improvement. 

M., R., and My. were practiced in the singing of the high tone. 
M. and R. made consistent improvement in the amount of error. 
The figures given for My. seem to indicate a lack of capacity 
for training with the high tone of a similar nature to that found 
in the case of A. with the low tone. The similarity, however, 
does not hold in one respect. My.'s results were, indeed, erratic 
during the course of practice but he frequently approximated the 
standard to a much higher degree than did A. This was espe- 
cially true of that part of the practice series during which the 



DISCRIMINATION AND SINGING OF TONES 175 

singing took place simultaneously with the sounding of the stand- 
ard fork. Indeed, it was true of all the subjects except A. that 
a marked improvement took place during this part of the practice 
series. In the case of My., however, this improvement was not 
carried over when the original conditions were restored, while in 
the other cases it was. 

At the end of the practice series, then. F., Ay., M. and R. had 
made improvement in the singing of tones of the pitch on which 
they had practiced. At this stage the average errors, respectively, 
for these subjects amounted to 2.3%, 0.6%, 0.6%, 0.7% of the 
standards. 

Turning now to a consideration of the effects of the practice 
on the singing of tones at a different level, it will be seen that of 
those who improved in the practice series F., M. and R. made 
a slightly smaller amount of error in these non-practiced tones 
after practice than before practice, while Ay. did worse. There 
is, therefore, little evidence of transference of practice effects in 
the accuracy of singing from one pitch to another. 

The direction of the error represented by the figures in column 
Vj is not indicated in the table, these figures being the arith- 
metical average of errors. A detailed study of this point shows, 
however, that there is a general tendency to sing the lower tone 
too high, and the higher tone too low. This statement is un- 
equivocally true of the tones sung by A., F., M. and Ay., both 
before and after practice. In the case of R. both the low tone 
and the high were sung too high before and after the practice 
series. My. always sang the low tone too high but sang the 
high tone too low before practice and too high after practice. 
Miles (7) found a general tendency to sharp in male subjects 
but the fact that his standard tones covered a narrower range 
within easy compass of his subjects' voices made his results 
scarcely comparable with our own. 

Seashore and Jen tier (9) report improvement in accuracy of 
singing tones after practice with visual control by means of the 
Seashore tonoscope. In these experiments the improvement ap- 
parently transfers to tones of a different pitch from those prac- 
ticed. That there should be more evidence of transfer under 



176 EDWARD HERBERT CAMERON 

such conditions where the visual control gives a knowledge of 
results might be expected from other experiments in transfer 

(4), (6), (8). 

A comparison of columns V 2 and V 3 before and after practice 
shows improvement in approximation to uniformity of tone both 
for the tones which were practiced and those which were not 
in almost all cases. The only exception is in the case of My., 
all of whose tones were less uniform after practice than before. 
The improvement with respect to uniformity is not marked in 
the case of A. and also in the case of M.'s high tones so far as 
the extreme variations at the beginning of the tone are con- 
cerned (V 3 ). 

Berlage (1) has found that the deviations from period to 
period are less when the reproduced tone is sung in response to 
one's own sung tone than when the standard tone is the sung 
tone of another person. He also found that the deviations are 
greater when the tone is sung freely without any attempt to ap- 
proximate a standard than it is under either of the conditions 
just mentioned. It seems, therefore, that the steadiness with 
which a tone is maintained is not a purely physiological phenome- 
non of muscular tetanus but that it varies with the psychological 
conditions and is subject to improvement through practice. 
Furthermore, the effects of practice, so far as steadiness is con- 
cerned, are transferred to tones of a different pitch from those 
practiced, as my present investigation shows. The same state- 
ment applies to the larger deviations at the beginning of the tone. 

It remains to give the results of the tests for discrimination 
after the practice series of experiments. These tests were given 
in the same way as those made before practice. The results are 
given in Table V and the average results before practice are 
added for purposes of comparison. 

This table shows that F., Ay., M. and R. have lower thresholds 
of discrimination at the level at which practice in singing took 
place than was the case before practice. This improvement is, 
however, not extended to the other non-practiced level. It will 
be remembered that these subjects all made unequivocal improve- 
ment in the amount of error made in singing their standard 



DISCRIMINATION AND SINGING OF TONES 

Table V 
Discrimination-Thresholds after practice 



1/7 



Low 






High 








A. 


F. 


Ay. 


M. 


R. 


My. 


A. 


F. 


Ay. 


M. 


R. 


My. 


i. 


2.4 


•9 


1.3 


1.9 


1-3 


2.4 


3.8 


2.0 


2.0 


1.1 


1.1 


2.0 


2. 


2.4 


1.3 


1-3 


1.9 


1-3 


1.9 


3-8 


2.6 


2.0 


0.5 


1.5 


2.6 


3. 


2.4 


0.9 


1-3 


1.3 


1-3 


1.9 


2.6 


2.0 


2.0 


0.5 


i-5 


2.6 


4. 


1.9 


0.9 


1-3 


1.9 


0.9 


1.9 


3-2 


2.6 


2.0 


0.5 


1.5 


2.0 


5. 


2.4 


1.3 


1-3 


1.9 


1-3 


1.9 


3-2 


2.6 


2.0 


0.5 


1-5 


2.0 


Av. 


2.3 


1.1 


1.3 


1.8 


1.2 


2.0 


3-3 


2.4 


2.0 


0.6 


1.4 


2.2 


Before | 


1 


] 




















Practice 


2.2 


2.2 2.2 


14 


1-3 


1.8 


3-,4 


2.6 


2.1 


1.8 


2.6 


2.1 



tones. In the case of A. and My. no improvement is shown in 
the pitch discrimination after practice in singing. As has already 
been shown these subjects made no real progress in the reduction 
through practice of the error in singing the standard tone. 

Comparing tables IV and V it is found that in general the 
figures expressing sensory discrimination are lower than the 
corresponding figures for the sung tones. This is true in all but 
three of the twenty-four pairs of results. In the three cases 
where this is not true, the figures are practically the same. It 
would, therefore, appear that fineness of discrimination is more 
accurate than motor reproduction. This point becomes very 
marked when it is remembered that the figures given for sensory 
discrimination are probably much higher than would have been 
obtained from tests of sensory discrimination given under the 
more standardized threshold conditions. 

An inspection of the figures would lead us to expect little cor- 
relation between sensory discrimination and accuracy of repro- 
duction for the same tone. Calculated by the Pearson pro- 
ducts-moment method the index of correlation is .37 with a 
probable error of .023. Miles (7) found the index of correla- 
tion between accuracy of singing and pitch discrimination for 
eighty- two male subjects to be only .33. The most marked lack 
of correlation between accuracy of reproduction and sensory 
discrimination is in the cases of A. and F., the subjects who made 
little or no improvement in accuracy of singing but who never- 
theless are not noticeably different from the others in the results 
of the discrimination tests. That practice results in a lowering 
of the threshold of pitch discrimination has been reported by 



178 EDWARD HERBERT CAMERON 

Seashore (10) and also Smith (11). Seashore distinguishes a 
so-called "cognitive" threshold from a "physiological" threshold 
and holds that the cognitive threshold only is subject to improve- 
ment through practice. It is clear, however, that the physio- 
logical threshold is a somewhat theoretical limit. It must be 
admitted, however, that the threshold found in this investigation 
is a "cognitive" threshold found under special conditions. The 
subjects were required to be certain not only of a difference in 
pitch but of the direction of the change. While this kind of 
cognitive threshold, therefore, was peculiarly subject to improve- 
ment by practice, it seems nevertheless true that the improvement 
which actually took place was due to the practice in singing 
rather than to adaptation, attention, interest, or other factors. 
Such factors would presumably be operative to as great a degree 
at the non-practiced level as at the level at which practice took 
place but, as we have seen, the improvement was not transferred 
from one level to another, either in the singing or sensory dis- 
crimination. 

The results of the present study tend to confirm the objections 
that have been made to those so-called motor theories which re- 
gard kinaesthetic sensations as playing the fundamental role in 
such responses. The results show that motor reactions to tones 
are by no means so accurate as the sensory discrimination. Judd's 
study of the eye-movements in the perception of the Miiller-Lyer 
illusion (5) gave analogous results. He found that there was 
an intimate relationship between the character of the eye move- 
ments and the amounts of illusion, and also that both the eye- 
movements and the amount of illusion changed with practice. 
Nevertheless, the movements are of such a nature as to preclude 
the probability that the resultant kinaesthetic sensations are the 
cause of the illusion. 

On the other hand, there is agreement between these results 
and those of other previous investigations which have shown 
the closeness of the relationship between sensory and motor 
processes — a relationship involving motor organization rather 
than mere kinaesthetic sensations. In my earlier investiga- 
tion it was shown that the motor reactions to tones are 



DISCRIMINATION AND SINGING OF TONES 179 

intimately related to all the forms of harmony and dis- 
cord. Bingham's investigations (2) have led him to the con- 
clusion that melody is related to the "upsetting of established 
muscular tensions," "the organization of incipient responses," 
and the merging of balanced tensions. Stetson (12) has found 
similar tension-relaxation processes in his experimental study 
of rhythm. The facts here presented point to the organic unity 
of motor and sensory factors in even so relatively simple a process 
as sensory discrimination of tones. With the development of 
more precise and unvarying modes of response to one tone, there 
arises a greater keenness in discriminating that tone from all 
others. Sensory discrimination must, therefore, be regarded as 
related to the organization which has taken place with reference 
to the new mode of response. 

SUMMARY OF RESULTS 

(i) There is no marked correlation between the initial capa- 
cities of the subjects tested for discrimination of tones and ability 
to reproduce these tones accurately by singing. A subject (A) 
whose "ear" is little inferior to that of the other subjects is never- 
theless totally at a loss to sing the tones accurately. 

(2) Practice in singing tones of a certain pitch resulted in 
marked reduction in the error of reproducing those tones in the 
case of four of the six subjects. 

(3) Slight improvement in singing tones of a pitch different 
from the one practiced was made by three of those four subjects 
and no improvement by the other. 

(4) Practice resulted in improvement in steadiness both at 
the initial point of the sung tone and throughout the tone as a 
whole in the case of those subjects who improved in accuracy. 

(5) Improvement in steadiness was also shown in the singing 
at the non-practiced level. 

(6) Subjects who improved in accuracy of singing tones of a 
certain pitch improved also in discrimination of tones at that 
level. 

(7) There was no improvement for such subjects in the dis- 
crimination of tones of a different pitch from that practiced. 

(8) Subjects who did not improve in accuracy of singing 
made no improvement in discrimination. 



180 EDWARD HERBERT CAMERON 

REFERENCES 

1. Berlage, F. Der Einfluss von Artikulation und Gehor beim 

Nachsingen von Stimmklangen. Psychol. Stud., 19 10, 

6, 39-I40. 

2. Bingham, W. V. Studies in melody. Psychol. Rev., Monog. 

Sup pi., 1910, 12, 1-88. 

3. Cameron, E. H. Tonal reactions. Psychol. Rev., Monog. 

Sup pi., 1907, 8, 227-300. 

4. Judd, C. H. Practice and its effects on the perception of 

illusions. Psychol. Rev., 1902, 9, 27-39. 

5. Judd, C. H. The Muller-Lyer illusion. Psychol. Rev., 

Monog. Suppl., 1905, 7, 55-81. 

6. Judd, C. H. Practice without knowledge of results. 

Psychol. Rev., Monog. Suppl., 1905, 7, 185-198. 

7. Miles, W. R. Accuracy of the voice in simple pitch singing. 

Psychol. Rev., Monog. Suppl., 1914, 15, 13-66. 

8. Ruediger, W. C. Improvement of mental functions through 

ideals. Ed. Rev., 1908, 36, 364-371. 

9. Seashore, C. E., and Jenner, E. A. Training the voice by 

the aid of the eye in singing. /. of Educ. Psychol., 1910, 
1, 311-320. 

10. Seashore, C. E. The measurement of pitch discrimination : 

a preliminary report. Psychol. Monog., 1910-11, 13, 
21-60. 

11. Smith, F. O. The effect of training in pitch discrimination. 

Psychol. Rev., Monog. Suppl., 1905, 7, 67-103. 

12. Stetson, R. H. A motor theory of rhythm and discrete suc- 

cession. Psychol. Rev., 1905, 12, 250. 



AN EXPERIMENTAL STUDY OF THE CONSCIOUS 
ATTITUDES OF CERTAINTY AND UNCERTAINTY 

By John Trumbull Metcalf, Ph.D. 

Instructor in Psychology 
Smith College 

The object of the present investigation is an analysis of the conscious atti- 
tude (B ewusstseinslage) on the motor side. The effort is made to determine 
whether certain regular forms of bodily reaction accompany the sub- 
jective attitude of certainty, and whether with changes in subjective attitude 
there are corresponding changes in the forms of such reaction. The subject 
is given a task which he carries through in response to certain instructions. 
The motor processes immediately involved in carrying through this reaction 
to the instructions are recorded by the apparatus, and at the end of each 
experiment the subject gives a complete introspective report. These two 
records, the objective record of the form of the reaction and the introspective 
record of the conscious processes experienced during the reaction, are ex- 
amined together to see if there is any correspondence between them. The 
tasks chosen for the investigation consist in the making of drawing move- 
ments with the hand out of sight. The apparatus gives records of the ac- 
curacy of these movements, and of all changes in rate and pressure which 
occur during the process. Variations are introduced to determine whether 
by changing the instructions in such a way as to modify the objective form 
of the reaction a corresponding change may be produced in the subjective 
attitude. 

The conscious attitude has come to hold a very important 
place in psychological theory, and a good deal of work has been 
directed toward analyzing it by the introspective method. Yet, 
to the writer's knowledge, no attempt has been made to attack it 
on the motor side. The present investigation attempts this, its ob- 
ject being to find out whether the introspectively given attitude 
of certainty involves a definite form of motor reaction on the 
part of the subject, and a change in the attitude a change in the 
form of such reaction. The method used is an extension of that 
which has already been employed in the investigation of the 
relation between motor processes and consciousness. Judd (2) 
has shown that the character of perception and attention is closely 
connected with motor processes. The present work follows the 



182 JOHN TRUMBULL METCALF 

same method, extending it to that phase of consciousness which 
experimenters in the field of the higher thought processes have 
called the "Bewnsstseinslage" or conscious attitude. This in- 
vestigation is, therefore, more closely related to the experimental 
work on the relation between movement and consciousness than 
it is to the almost exclusively introspective experimental investi- 
gations of the higher thought processes. 

The first requisite in an investigation of this sort is a motor 
process, simple enough to be measured objectively, and yet suffi- 
ciently difficult of performance to yield the necessary richness 
of introspective data. The making of simple drawing move- 
ments with the hand hidden from sight was chosen as the task. 
These drawing movements were measured by the apparatus, and 
after each drawing full introspections were given by the subject 
and taken down by the experimenter. The drawings were made 
in response to certain instructions, and in the course of carrying 
out the process various attitudes appeared, as reported in the 
introspections. Now if there is any relation between attitude 
and motor processes in general the attitudes reported here must 
be connected with the particular motor processes measured by the 
apparatus, because both attitude and the processes measured come 
of necessity in response to the instructions. In other words, the 
motor processes, i. e., the drawing movements, measured by our 
apparatus are necessarily involved in carrying out the instruc- 
tions, and are therefore particularly relevant to any attitudes 
consequent upon the instructions. 

The apparatus for measuring the drawing movements was that 
used by Freeman in his analysis of writing movements ( i). The 
two cuts of the apparatus which appear there are reprinted here 
(Figs, i and 2). 

This apparatus yields three objective records of the drawing 
movements. In the first place there is the actual drawing made 
by the subject upon the primary sheet. The task was usually the 
copying of a simple geometrical figure — the "model." Thus 
when the drawing is measured for straightness of lines, length 
of lines, size of angles, etc., and these measurements compared 
with the corresponding measurements of the model, a record of 



rs &£-*.££ 




e» <U O «■> 



CD 
-£ tft "Q 

55 <D j-j CO -M 

■a o- g 






* > 

O (L) +j 

g a - s 



o 



i8 4 



JOHN TRUMBULL METCAiL 




Fig. 2. 

accuracy is obtained. In the second place, the apparatus gives 
a time record for the whole drawing or any part of it. As the 
subject draws upon the primary sheet his drawing is registered 
through the typewriter ribbon upon the moving strip. This gives 
a record of the whole drawing spread out under a standard time- 
line traced by the first marker. By correlating this with the 
original drawing in the manner described by Freeman, it is possi- 
ble to follow the details of time changes throughout. In the 
third place, the apparatus gives a record of pressure. The draw- 
ings are made above the hinged plate which sinks slightly when 
pressure is exerted upon it. This movement of the plate, much 
magnified, is communicated to a pointer which traces a curve 
upon a smoked strip. In this way all changes in the pressure 
exerted by the subject in drawing are registered. This pressure 



EXPERIMENTAL STUDY OF CONSCIOUS ATTITUDES 185 

record is correlated with the records of time and accuracy 
through a second standard time-line traced upon the smoked 
strip by a second marker in circuit with the other. The time is 
given by a Kronecker Interrupter located in another room, 
marking tenths of a second. Three records are, therefore, ob- 
tained from each drawing movement, showing its accuracy, the 
rate of drawing throughout, and synchronous changes in 
pressure. 

Several changes were, however, made in Freeman's apparatus. 
A finely-pointed lead-pencil was used in place of his capillary 
pen for the time-record on the moving strip. An electric motor 
was used to drive the apparatus. This motor was encased in a 
wooden box heavily lined with felt to shut in the noise. Its speed 
was reduced through a worm-gear mounted in a standard on the 
top of the box. The drum, C, Fig. 1, was turned by a string-belt 
from the driving-wheel of the worm-gear. This method lacks 
only one of the advantages of the original arrangement. It is 
not now possible to start the apparatus at full speed, as could 
be done when the friction clutch was used. This proved to be 
no detriment, however, as it took the apparatus only a very short 
time — about two seconds — to develop its maximum speed. This 
time was just about the interval the experimenter allowed between 
the starting of the motor and the giving of the signal to begin 
drawing. The chief advantage in using the motor is that it 
eliminates a great deal of distracting noise. 

A third and more important change in the original apparatus 
was the substitution of a new pressure-recording device for the 
tambours. The tambours were used in the first practice series, 
and were found unsatisfactory for the purposes of this experi- 
ment. The pressure in a drawing movement is usually less in 
amount and more gradual in its changes than that in a writing 
movement. This calls for a recording device which is more 
sensitive and which magnifies the record to a greater degree. 
There is always a stretch to the rubber of the tambours and to 
the walls of the connecting- tube. The movements are thus 
softened down so that some of their characteristics are lost. The 
simple device used in these experiments does away with this de- 
ficiency, as well as satisfying the other conditions. 



•i86 



JOHN TRUMBULL METCALF 



03 

6 




EXPERIMENTAL STUDY OF CONSCIOUS ATTITUDES 187 

The two tambours, I and K, Fig. 1, were removed, as was also 
the metal disc on the end of the lever, F, Fig. 2. A new record- 
ing-pointer was made. As this pointer was of considerable 
length, the drum D, Fig. 1, over which the smoked strip runs, 
had to be moved away from the rest of the apparatus. It was 
accordingly placed about a meter away from the drum, C, and 
run by a string-belt from a wheel on the axle of the drum C, to 
a wheel of the same size on its own axle. The marker, L, Fig. 1, 
was held by a special standard which stood on the table. The 
new recording-pointer was made of a thin piece of bamboo, L, 
Fig. 3, 60 cm. in length. At the recording end it was supplied 
with a metal recording-point. Near the other end it was pierced 
by an axis, A, which was then mounted in bearings. The move- 
ments of the lever, F, Fig. 2, were communicated to the record- 
ing-pointer by means of a thread running over a system of 
wheels. These wheels were made of wood and were supplied 
with fixed axles mounted on bearings, friction being thus reduced 
to a minimum. Fig. 3 shows how the movements of the lever, F, 
are communicated to the recording-pointer, L, which is at right 
angles to it and on a different level. The wheels over which the 
thread, T, runs are numbered 1, 2, and 3. An end view of the 
arrangement of wheels 2 and 3 is given in Fig. 3a. The thread 
was attached to the recording-pointer between the axis and the 
point — 9 cm. from the axis and 42 cm. from the point. At this 
place too, on the other side of the pointer, there was attached a 
long fine spring, S, the tension of which could be readily and 
delicately regulated. A small round weight, W, Fig. 3, with a 
circular hole in the middle, was placed over the short end of the 
recording-pointer to make its weight equal on either side of the 
axis. 

Another addition to the apparatus was the screen for cutting 
off the subject's view of his hand as he drew. The screen was 
of grey card-board, held in place by means of clamps, which 
were in turn held by uprights clamped to the large surface, A, 
Fig. 1. Another large screen of black card-board was used to 
cut off the subject's view of the pressure apparatus. 

The marker, M, Fig. 1, used by Freeman to record the giving 



188 JOHN TRUMBULL METCALF 

of the signal in reaction experiments was not needed for our 
purposes, and was removed. A slight change was made in the 
arrangement of the marker, J. Instead of its being fastened to 
the plate, H, it was held just above it in a clamp attached to a 
rod running from one of the uprights holding the screen. To 
this same upright, higher up, were also attached the rod and 
clamp for holding the card on which the model was drawn. The 
models were drawn in black on white cards 12.5 cm. x 7.5 cm. 
The figures used are illustrated in Fig. 4, and will be described 
in detail in the statements of the individual series. The method 
of measurement adopted in each case will also be described there. 

The apparatus was run more slowly than in Freeman's experi- 
ments, drawing movements being made more slowly and with 
more gradual changes in speed and pressure than writing move- 
ments. Moreover, the experiment is of such a character that it 
does not demand as minute an analysis of the movements as did 
Freeman's. Accordingly the apparatus was run just fast enough 
so that the records could be read in iooths of a second. 

In an investigation of this sort another requisite is that the 
subjects give full introspections. In our experiments they were 
instructed to tell after each experiment everything that had gone 
on in consciousness from the giving of the signal to begin the 
drawing to the end of the drawing. The introspective reports 
were not confined to this period, however, and general remarks 
made by the subject were also accepted by the experimenter 
though they were not given the same weight as the introspective 
report. Sometimes the experimenter increased the fulness of 
the report by appropriate questions, but care was taken that these 
should not be suggestive. Those who served as subjects were 
all trained psychologists. They were Professors Angier (A.) 
and Cameron (C), Dr. E. P. Frost (F.), Instructor in Psychol- 
ogy, and Mr. S. L. Reed (R.)> a graduate student in Psychology. 
To these men I am indebted not only for the large amount of 
time they have given me as subjects, but also for many pieces of 
helpful criticism and advice. The subjects were kept as naive 
as possible. With the exception of A., none of them knew the 
purpose of the experiment, or anything about the results. A. 



EXPERIMENTAL STUDY OF CONSCIOUS ATTITUDES 189 

knew the purpose of the experiment, having himself suggested 
the problem, but he entered upon the investigation unconscious 
of any bias, and without anticipating any of the results. He was 
also kept in ignorance of his own results, though he did examine 
some of the results obtained from other subjects in cases where 
their task had been different from his own. 

The method of procedure was as follows. Before each experi- 
ment the model was placed in its holder. The subject seated him- 
self before the apparatus, took the pencil, and placed his hand 
under the screen ready to draw. He then received his instruc- 
tions. The experimenter, after a preliminary "Ready" signal, 
started the motor and the two time-markers and, finally, gave 
the signal "Now" for the subject to begin drawing. This suc- 
cession — "Ready," starting of apparatus, and "Now" — was not 
carried through in connection with any time-regulating mechan- 
ism, but the intervals were kept as nearly equal and constant as 
possible. As soon as the drawing was finished, the experimenter 
stopped the apparatus and took down the subject's introspections. 

The instructions in these experiments, though not given in a 
stereotyped formula each time, were nevertheless definite and as 
brief as possible. At the beginning of a series the experimenter 
described to the subject in exact terms how the figure was to be 
drawn, i. e., that he was to begin at a certain definite point, pro- 
ceed in a certain direction, and make his drawing as nearly like 
the model in size and form as he could. These instructions were 
briefly repeated for each drawing of the series. In all the earlier 
experiments no instructions were given concerning the rate of 
drawing or the pressure to be exerted, since variations in these 
objective factors, rate and pressure, were to be correlated with 
variations in introspective attitude, to give instructions about 
them would be to predetermine one of the variables. Later on, 
however, special instructions as to rate, and in a few cases as to 
pressure, were given in order to see if by purposely varying these 
objective factors a corresponding variation in attitude could be 
obtained, such as preceding experiments might have led one to 
expect would occur. When such special instructions were given 
they were always repeated in exactly the same words before each 
drawing. 



igo JOHN TRUMBULL METCALF 

At the start the subjects were told to report in their introspec- 
tions all conscious processes that accompanied the drawing of the 
figures. No hint was given as to what processes were the object 
of the investigation, and all attitudes reported were given equal 
consideration. Early in the investigation, however, it was found, 
as had been to some extent anticipated, that the only attitudes 
that appeared universally enough to admit of systematic study 
and correlation were attitudes designated as "certainty" and "un- 
certainty." It was therefore to these attitudes that the experi- 
menter directed his attention, and the variations introduced into 
the experiment were aimed directly at bringing them more clearly 
into the light. Moreover, the questions by which the experi- 
menter enlarged the subject's introspective report often had to 
do directly with the attitude of certainty, although they did not 
reveal to the subject the fact that this attitude was the object of 
the investigation. The subjects still continued to report all the 
other attitudes, images and sensations they had experienced. It 
may be that other observers would distinguish other attitudes 
within the certainty and uncertainty experienced in a drawing 
process. However this may be, our subjects reported definite 
conscious attitudes, some of which were designated as attitudes 
of certainty and others as the opposite attitudes of uncertainty, 
and the contrast between them holds whether or not they might 
be called by other names or each divided into a number of dif- 
ferent sub-classes. Certainty, however, may be of two kinds, as 
reported by our subjects. These are designated as positive cer- 
tainty and negative certainty. Positive certainty is the attitude 
that the drawing made is correct. Negative certainty is the atti- 
tude that the drawing made is incorrect in some definite way. Un- 
certainty is the attitude that the subject can't tell whether the 
drawing approximates the model or not. The distinction between 
positive and negative certainty involves a distinction between the 
attitude of certainty and the accuracy with which the subject 
feels the drawing to have been made. The attitude of certainty 
might be that the drawing was accurate or it might be that it was 
inaccurate. In either case the attitude was certainty. 

Certainty in a drawing may give way to uncertainty and vice 



EXPERIMENTAL STUDY OF CONSCIOUS ATTITUDES 191 

versa. Sometimes the attitude refers to one portion of the draw- 
ing, sometimes to the drawing as a whole. In the longer draw- 
ings there is reported a continuous attitude, which lasts 
throughout, varying between certainty and uncertainty. In 
shorter drawings the attitude may appear just as the subject 
finishes, and in such cases it usually refers to the drawing as a 
whole. There are, moreover, different grades to the attitude. 
At times certainty or uncertainty is more intense than at other 
times. These differences are reported by the subject in such 
terms as "good certainty," "only fair certainty," "a little uncer- 
tain," "typical uncertainty," "absolutely uncertain," etc. 

In order to show a correlation between the attitudes reported 
and the objective records, it is evident that a fairly large number 
of experiments must be made, for the safety of this correlation 
depends upon the number of experiments and the taking of care- 
ful introspections. With the more complicated models that were 
used first in our experiments, twenty drawings were made of 
each model. With the simpler drawings, ten experiments were 
usually obtained for a series. In all about six hundred and fifty 
experiments were made. 

We turn now to a consideration of the individual series of 
experiments. 

Series I. Practice Experiments 

The first experiments were conducted in the spring of 19 12, 
chiefly for the purpose of finding out whether the apparatus and 
method were adequate. As already stated, these preliminary ex- 
periments showed that a new method of recording pressure would 
have to be devised and substituted for the tambours. 

The drawing process in this series was the same for all sub- 
jects, the model being a single straight, vertical line, 27 mm. in 
length (Fig. 4, No. 1). The task was to draw three vertical 
lines, equal in length to the model line, parallel to each other, 
and equal distances apart. They were to be drawn from top to 
bottom, and the whole process to progress from left to right. 
At the close of the drawing the subject gave his introspections, 
and named the order of certainty of the lines he had drawn. He 



192 



JOHN TRUMBULL MET CALF 






CO 




to 



vA 




«o 



EXPERIMENTAL STUDY OF CONSCIOUS ATTITUDES 193 

ranked them from most certain to least certain. After a few 
experiments had been made, it was found that the first of the 
three drawn lines came to be regarded, without intention, as a 
sort of preliminary affair in which the subject simply "got the 
swing" of the process. The result was that he had little intro- 
spective knowledge of it, and it could not be compared with the 
other two lines. Since the introspective comparison of three 
lines was desired, this difficulty was obviated by having the sub- 
ject draw four, the first of these to be a preliminary one to get 
the process started. 

The lines in the drawing made by the subject were measured 
for accuracy as follows. First, their length was measured in 
millimeters. Second, an estimate of the straightness of each line 
was made. This was expressed in two ways. First the number 
of little waves or turns in direction was noted, and second the 
amount of deviation of the whole line from the vertical — de- 
termined with reference to a perpendicular to the horizontal 
boundary of the bottom of the space in which the subject drew. 
The distance from this perpendicular of the top of the line was 
measured and compared with a similar determination of the bot- 
tom of the line. Their difference, expressed in millimeters, gives 
the second measure of straightness. Third, the width of each 
interspace was calculated. This was done by taking the average 
distance apart of the two lines bounding it. 

For purposes of time measurement the lines were divided into 
three equal parts and the time for each third calculated. In 
nearly every line, moreover, there were two periods, one at the 
beginning and another at the end, during which the pencil was 
touching the paper but not moving. The subject was not aware 
of the fact that he did not begin to draw the instant his pencil 
touched the paper, nor did he know that at the end of the line 
he held it still for a fraction of a second before raising it. The 
time for these two periods was also determined, making in all 
five time determinations for each line. Finally, the time between 
the end of each line and the beginning of the next, or the time in 
the interspaces, was also calculated. 

These practice results are incomplete, the pressure records 



194 JOHN TRUMBULL METCALF 

being defective, and are hardly significant except as they point 
toward later results. The records of the lines which each sub- 
ject called most certain, and those for his least certain lines were 
grouped together and averaged. In the cases of A. and R. the 
average most certain line is drawn at a slightly faster rate than 
the average least certain line. In the case of C. the opposite is 
true here, but later results do not bear this out. With F. it was 
impossible to compare the most certain and least certain lines, 
because of lack of complete introspection. 

Certain differences in imagery between the individual subjects 
came out in the practice series, and, as they keep appearing 
throughout the whole investigation, it will be well to describe 
them here. 

A.'s imagery is almost entirely visual. In starting a drawing 
he has a visual scheme of it to be filled in. This visual scheme 
is followed throughout, and any marked deviation from it is 
pictured in visual terms. Sometimes the visual scheme is weak 
and the imagery vague. In making a drawing without a model 
it seems important for this subject to keep his eyes fixated upon 
the cardboard screen in the general direction of his hand. In 
drawing from a model he keeps his eyes upon the card, but his 
tendency is, as he expresses it, "to use the model as a suggestion 
rather than as a plan," that is, to draw rather from his own 
visual image of the figure than from the figure itself. 

C. reports chiefly kinaesthetic imagery and sensations in the 
drawing process. Any plan or scheme he has at the beginning 
of a drawing is in kinaesthetic terms. He sometimes prepares 
for a drawing by making incipient movements with the pencil. 
He is easily disturbed by any slight awkwardness in the manner 
of holding the pencil, or in the position of the hand. In draw- 
ing from a model he follows it with his eyes throughout. In 
drawing without a model he does not keep his eyes fixated in a 
particular way, but gets his hand placed carefully and then looks 
away. He reports visual imagery too, but his visual image al- 
ways comes up after the drawing has been made, as the last stage 
of the process. 

F. reports imagery of both types. The most prominent kind 



EXPERIMENTAL STUDY OF CONSCIOUS ATTITUDES 195 

of imagery with him is visual-motor. He images the developing 
figure as he draws, and he often visualizes the moving pencil- 
point as well. 

R. reports less imagery than any of the other subjects. What 
he does report is chiefly visual, with occasional kinaesthetic, 
imagery. Often incomplete visual images appear. Part of a 
line will be visualized and the rest will not appear. Sometimes 
a line will be visualized as heavy and dark at one end, but grad- 
ually shading off into nothing at all at the other end. 

Series II, IIa and IIb 

The remainder of the experiments were made in the fall of 
1912, and the winter and spring of 1912-13. In the present 
series each subject was given a different figure to draw, and so 
a different task. Series IIa is in each case a variation upon Series 
II. In treating these series we will take up the different subjects 
separately. 

Subject A. — A. was given as a model a circle, 27 mm. in 
diameter (Fig. 4, No. 2). He was instructed simply to copy 
the circle exactly, beginning at the top and drawing in the anti- 
clockwise direction, using a finger movement and not shifting the 
position of his hand during the drawing. The anti-clockwise 
direction was chosen because it seemed to the subject that it 
would be more natural for him than the clockwise direction. 
After taking a series of twenty such anti-clockwise drawings 
a variation was introduced by having the subject draw the circle 
in the opposite direction, still beginning at the top. A series of 
twenty such clockwise drawings was also obtained, forming 
Series IIa. Then, after a long interval (October to April), dur- 
ing which a large number of other experiments had intervened, 
another series of ten anti-clockwise drawings was made. These 
form Series lib. 

The method of measurement for these drawings was as fol- 
lows. For accuracy, the approximate center of the circle drawn 
by the subject was found with a compass. Then four axes were 
drawn through it, the vertical, the horizontal, and the two diago- 
nal axes. When compared with the diameter of the original 



196 JOHN TRUMBULL METCALF 

circle, the lengths of these axes give a good estimate of the ac- 
curacy of the circle as to size. When compared with each other 
they give a good measure of the roundness of the circle. In 
measuring the accuracy of the circle, also, the distance from end 
to end, that is, the amount by which the end point failed to meet 
the starting-point, was measured in millimeters. This measure- 
ment is expressed as the "distance from end to end" under the 
abbreviation "E. to E." 

For measuring time changes during the process of drawing, 
the circumference was divided into quarters. A more minute 
division of the circumference seemed unnecessary because of the 
character of the subject's introspections. These were nearly al- 
ways given in terms of halves or quarters of the circle, the sub- 
ject making these divisions guiding marks in his description of 
the course of mental processes during the drawing. It was 
therefore evident that any changes in drawing rate which could 
be correlated with a change of attitude should show in a com- 
parison of the quarters. The end-points of the vertical and hori- 
zontal axes were taken as marking the boundaries of the quarters 
of the circumference. This, of course, is not absolutely accurate, 
as the drawn circle was never perfectly round, but it is exact 
enough for our purpose. The pressure record was also divided 
into four quarters. 

For brevity and convenience in exposition, the results of these 
series will be treated as a whole by a consideration of the aver- 
aged measurements of the twenty drawings and a summary of 
the introspection. The individual drawings show such uniform- 
ity in their characteristics that this may be done. 

The introspections for both Series II and Ha show that a con- 
tinuous attitude accompanied the drawing of the circle from start 
to finish. The anti-clockwise circles are more certain than the 
clockwise. This is shown both in the relative number of atti- 
tudes of certainty reported, and in the general remarks of the 
subject. In both types (clockwise and anti-clockwise) the draw- 
ing of the second half is accompanied by less certainty than the 
first. In Series II, the anti-clockwise circle, a point of maximum 
uncertainty is sometimes located, and when it is thus mentioned, 



EXPERIMENTAL STUDY OF CONSCIOUS ATTITUDES 197 



as it is in six of the experiments, it is always located as either 
in the third quarter of the circle or between the third and fourth 
quarters, at the place where the figure III appears on a clock-face. 
In Series Ha (clockwise) a point of maximum uncertainty is 
sometimes mentioned; but it does not show any uniformity in 
the place at which it occurs, except that it is usually somewhere 
in the second half of the drawing. 

The accuracy and time measurements for Series II and Ha 
are compared in Table I. 









Axes 




Table I 
Subject A 




Time, in 


seconds 






II 


Ha 


II 


la 


No. 


Lgth. 


M.V. 


Lgth. 


M.V. 


Quarter 




M.V. 


M.V. 


1. 
2. 

3- 

4- 


21.8 
24. 

23.S 
21.6 


2-35 
2.35 
2.69 
3.12 


21. 1 

21.5 

21.5 
21.9 


2.01 , 

2.72 

2.S7 

2.25 


1st 

2d 

3d 
4th 


2.91 

2.45 
1.54 
3-26 


.62 
.40 
.30 
•43 


3-39 

2.73 

2-57 
2.91 


.58 

.31 

.40 

•45 


Ave. 
E. 


22.8 
to E. 




21.5 
II. 


6. 


Total 
Ha. 3-7 


10.16 


1. 41 


11.60 


1.07 



It will be seen from this table that the circles are very nearly 
the same size. Their average diameters differ by only 1.3 mm. 
They differ, however, in shape, the anti-clockwise circle being 
slightly elongated along its first diagonal axis. The clockwise 
circle shows less difference in the lengths of its axes, and is more 
nearly round. The distance from end to end is less too. Thus, 
the anti-clockwise circle is a little more accurate than the other 
in size, while the clockwise one is slightly more accurate in shape. 
So the less certain circle is a little less accurate than the other in 
size, but more accurate in shape. 

We turn now to the consideration of the time records. It 
will be seen from the table that the four quarters of the circle 
vary a great deal from each other in the speed with which they 
are drawn. In the anti-clockwise circle the drawing begins slowly 
in the first quarter, becomes somewhat faster in the second, is 
fastest in the third, and then in the fourth is slower than ever 
before. Thus the maximum change of speed comes between the 
third and fourth quarters, and the change is a retardation. In 
the other circle we find smaller differences between the times 



ig8 JOHN TRUMBULL METCALF 

of the respective quarters. The third quarter is again the fast- 
est, but the difference between it and the fourth is not as great. 
Instead, the greatest difference is between the first and second 
quarters, but this too is smaller than that between the third and 
fourth quarters of the other circle. There the difference was 
1.72 sec, here it is only .66 sec. Moreover, the change in the 
first case is a decrease, here it is an increase in speed. In con- 
nection with these characteristics of the time records it will be 
remembered that in the anti-clockwise circle the point of maxi- 
mum uncertainty, when it was located, came in the third quarter 
or between the third and fourth. Also, that while less certain 
as a whole than the anti-clockwise circle, the clockwise one does 
not show any definite point of maximum uncertainty. Thus, in 
so far as we have been able to locate a point of maximum un- 
certainty, that point comes at the place where the greatest change 
in speed is taking place, and that change is a decrease. In gen- 
eral the clockwise circle is less certain than the anti-clockwise one, 
and the table shows that it is drawn more slowly. The second 
half of each type of circle is less certain than the first, and the 
table shows that in each case the first half is drawn at an in- 
creasing, the second half at a decreasing rate. The point of 
maximum uncertainty comes in the anti-clockwise circles at the 
point where the figure III is located on a clock-face. This is the 
same point with reference to the hand as the end of the first 
quarter in the clockwise circle. In the first case uncertainty is 
reported, in the second certainty. In the first case there is a 
greater change in rate and it is a decrease, whereas in the second 
case the change is smaller and it is an increase. 

The pressure records show certain similar characteristics that 
hold throughout. In the records for both circles there is a 
gradual increase of pressure at the beginning, which lasts as a 
rule throughout the first half, and sometimes as far as the middle 
of the third quarter. When the third quarter is reached there 
is usually a decrease in the pressure curve of the anti-clockwise 
circle. This decrease appears in fourteen of the twenty drawings 
of Series II. In two drawings it does not appear, and in the 
remaining four the records are defective. This decrease in pres- 



EXPERIMENTAL STUDY OF CONSCIOUS ATTITUDES 199 

sure comes at the same part of the circle as the great decrease 
in rate revealed by the time records. It is also at the place at 
which the maximum uncertainty is reported when it is reported 
at all. In the two cases where there is no decrease in the third 
quarter the introspections show that the circles are unusually 
certain ones. Thus, the decrease in rate and pressure between 
the third and fourth quarters seems connected with the appear- 
ance of uncertainty. In Fig. 5, No. 1, we have a tracing of a 
curve of this sort from Series II. This curve is reduced in its 
length but not in its height. The differences in the height of the 
original curve are so slight compared with its great length that 
they would not show well if the reduction were made propor- 
tional throughout. All the tracings in this article, however, ex- 
cept those in Fig. 5 and two in Fig. 6, are made proportional 
throughout. Reducing is done by means of a pantagraph. In 
all the curves a rise indicates an increase in pressure. 

The pressure curves of Series Ha, the clockwise circles, show 
much the same characteristics as the others in their first halves. 
There are, however, a somewhat larger number of changes in 
pressure. The second half, in six cases out of the twenty, shows 
a decrease between the third and fourth quarters. In three of these 
cases it is accompanied by an attitude of uncertainty. In Fig. 5, 
No. 2, is given a typical pressure curve for a clockwise drawing. 
It shows in its first half the same general characteristics as the 
curve for the other circle, but it has not the decrease between the 
third and fourth quarters. In Fig. 5, No. 3, we have one of the 
clock-wise circles which differed from the others of the series 
markedly. The subject said in his introspections on this circle 
that he started not with the intention of following the model 
around as he drew, but rather to draw the circle in one quick 
sweep. The time records show that this circle was drawn nearly 
twice as fast as the average, and there is less slowing up in the 
second half. The result was that the changes in attitude were 
reduced in degree. The difference in certainty between the first 
and second halves was less markd. The curve shows scarcely 
any difference from beginning to end. So we find that in this 
experiment where the typical difference in certainty is reduced, 



200 



JOHN TRUMBULL METCALF 



r~ 



V 



10 

6 
to 




c* 



EXPERIMENTAL STUDY OF CONSCIOUS ATTITUDES 201 



the difference in the character of the pressure curve for the two 
halves is also reduced. 

Summing up the results for Series II and Ha, we find that : — 

(1) The anti-clockwise circle, which the subject said at the 
start would be more natural for him, is found in general to be 
drawn with more certainty than the other one. The first half 
of each circle is more certain than the second half. 

(2) The anti-clockwise circle is more accurate in size, but less 
accurate in form than the clockwise one. 

(3) The clockwise circle is drawn at a slower rate than the 
anti-clockwise one. In each circle the first half is drawn at an 
increasing rate and the second half at a decreasing rate. 

(4) The pressure curves for the two circles are similar in 
that there is a gradual rise throughout the first half. The second 
halves of both curves differ from the first halves. In the anti- 
clockwise drawings there is a decrease of pressure between the 
third and fourth quarters. In the clockwise ones this decrease 
sometimes appears, but there is no regularity about it. 

(5) Attitudes of uncertainty go with the second halves of the 
circles, and especially with the middle of the second half of the 
anti-clockwise one. At these points there is a decrease in the rate 
of drawing, and, in the anti-clockwise circle, a decrease in 
pressure. 

At the end of all the experiments a series of ten anti-clockwise 
circles was made, Series lib. The results of this series appear 
in Table II, which follows the plan of Table 1. 

Table II Subject A 





Axes 






Time 




No. 


LRth. 


M.V. 


Quarter 




M.V. 


1. 
2. 

3- 
4- 


144 
151 

14.5 
14.4 


1.2 
1.1 
1.2 
1.2 


1st 

2nd 

3rd 

4th 


3-88 
2.89 
2.12 
3-30 


•51 

45 
.26 

-3S 


Ave. 


14.6 




Total 


12.19 


1. 12 



This table shows that the circle is now drawn more accurately 
as far as its shape is concerned, but it is made much too small. 
The subject gave no indication in the introspections that he was 
aware of the small size of his drawing. This variation was intro- 



202 JOHN TRUMBULL METCALF 

duced primarily to see if there would be any indication of prac- 
tice effect in the results, and what the effect of this would be upon 
the attitudes reported. The subject noted some indication of 
practice effect. This did not appear immediately, but was re- 
ported first in the fifth drawing of the series. 

In general the drawings of this series are more certain than 
those of Series II. The subject is able to tell more clearly just 
what he has done, he reports more attitudes of both positive and 
negative certainty, and fewer attitudes of uncertainty. There is 
no longer a specific place at which the maximum uncertainty 
appears. It is noted only once, and there it is in the fourth 
quarter. An examination of the time records shows that the 
drawings as a whole are now made more slowly than in the other 
series, and there is less difference in time between the third and 
fourth quarters. 

The pressure records of this series are not very satisfactory, 
owing to the very slight pressure with which the subject drew. 
Four of the ten are not good enough to give results. In the other 
six the pressure is fairly constant in its increase from beginning 
to end, or else it increases quickly at the beginning and maintains 
about the same level to the end. There is never a decrease be- 
tween the beginning and the end. One particularly certain draw- 
ing was obtained — one in which no uncertainty is reported any- 
where in the course of it. The pressure curve for this drawing 
is illustrated in Fig. 5, No. 4. It will be observed that it rises 
gradually throughout, with no marked changes, and never a 
decrease. 

In this series, therefore, we find that with the greater certainty 
revealed in the introspections there is a change in the time and 
pressure characteristics, this change being in the direction which 
we have already found to tend toward certainty. The circle as 
a whole, however, is drawn more slowly than before. Previous 
results have seemed to indicate that certainty goes with a fast 
rate of drawing. Thus it would seem that in a long drawing 
of this character the rate of the total drawing is less important 
for the attitude than relative rates of the different parts. 

Subject C. — With C, the model in Series II and Ha consisted 
of an equilateral triangle, the sides of which were 28 mm. in 



EXPERIMENTAL STUDY OF CONSCIOUS ATTITUDES 203 



length. This is No. 3 in Fig. 4. In Series II the subject was 
instructed to draw the triangle beginning at the apex, then draw- 
ing obliquely downward to the left, then horizontally to the 
right, and then obliquely upward to the left to the apex again. 
Twenty drawings were made in this way. In Series Ha the sub- 
ject began to draw the triangle at the lower left-hand corner, 
going first along the horizontal line to the right, then obliquely 
upward to the apex, and obliquely downward to the starting- 
point again. 

This figure proved to be the least satisfactory of all those used. 
The drawing of it is not really a continuous process like the draw- 
ing of the circle, because the subject pauses at the corners. On 
the other hand, the three lines are not repetitions of the same 
mechanical process, as are the three vertical lines of the practice 
series. That these characteristics are unfavorable to the appear- 
ance of contrasts in attitude which can be reported, is shown by 
the character of the introspections. There was no continuous, 
changing attitude as in the case of the circle. On the other hand, 
the lines of the triangle were not compared with each other as 
were the lines of the practice series. Thus, different parts of the 
same figure cannot be compared, and the only comparison the 
introspections allow is of one whole figure with another, or a 
certain part of a whole figure with a similar part of another. 

The method of measurement was to measure the lines and 
angles of the drawn triangle and the amount by which the end 
failed to meet the beginning. The lines are numbered 1, 2, and 3 
in the order drawn in each case. The angles are similarly 
numbered. 

Table III gives the accuracy and time records for Series II 
and Ila. The average of twenty drawings is given in each case. 
The length of the lines is given in millimeters, as is also the dis- 
tance E. to E. The angles are given in degrees. 

Table III Subject C. 



II 


No. 


Lgth. 


M.V. 


Angles. 


M.V. 


Time 


M.V. 


1. 
2. 

3- 


254 
364 
304 


37 
5-9 
3-3 


55-8 

70.9 
57-2 


1.8 

39 

2.6 


2.30 
2.67 
2.14 


.57 
.63 
•9i 




E. tc 


> E. 16.5 


Total 7.1 1 


i-39 



204 



JOHN TRUMBULL METCALF 



Ha 



No. 


Lgth. 


M.V. 


Angles. 


M.V. 


Time 


M.V. 


i. 

2. 

3- 


43-8 
347 
40.3 


5-1 
5-9 
3-3 


79-9 

47- 

50.3 


3-8 
47 
4-5 


5.40 

473 

4.90 


1.46 
1.69 
i-57 



E. to E. 14.5 



Total 15.03 



4.66 



The large mean variations shown in this table indicate that 
there was very little uniformity in the drawing in either series, 
especially in the time. This makes it impossible to draw any 
conclusions, and owing to the difficulty of getting definite intro- 
spective comparisons between well-defined parts of these draw- 
ings, it is not possible to derive much of value from the pressure 
records. The only possible correlation between certainty and 
pressure in Series II is a very broad one. Of the fourteen cases 
in w r hich drawings had been designated as certain as a whole or 
uncertain as a whole, five were called certain and nine uncertain. 
As a whole the pressure curves of the certain drawings are more 
smooth and even than those of the uncertain drawings. The 
changes in pressure are more gradual and less frequent in the 
certain drawings than in the others. The difference between the 
certain and uncertain curves is not very great to be sure, but it 
holds in all but two cases out of the fourteen. Fig. 6 gives in 
No. 1 a tracing of a typical curve for a certain drawing, and in 
No. 2 one for an uncertain drawing. These curves, as in the 
case of the circles, are so long that a reduction of their height 
in proportion to the reduction of their length would make the 
changes in height very small. They are therefore reduced more 
In length than in height. 

In Series Ha no such comparison of whole drawings is possible 
as the subject there gives detailed introspections of the process, 
but does not give an indication of an attitude toward the figure 
as a whole. There is, however, one very marked peculiarity 
which appears in some of the pressure curves. This is a sort of 
"plateau" at the end. It will be remembered that in this figure 
the end comes at the lower left-hand corner, and the last line is 
the downward diagonal from upper right to lower left. At the 
end of this line and completing the figure, this plateau occurs in 
eight out of the twenty cases, being absent in the other twelve. 



EXPERIMENTAL STUDY OF CONSCIOUS ATTITUDES 205 



\ 




6 



cv 



^ 



<o 



2o6 JOHN TRUMBULL METCALF 

In Fig. 6, No. 3, a marked plateau is illustrated. No. 4 is an 
ending of the same sort, but not as marked, and No. 5 illustrates 
the end of a curve in which there is no plateau. The reduction 
of these is uniform throughout. The introspections show that 
in each drawing where a plateau appears at the end of the pres- 
sure curve there was an attitude of certainty at the instant of 
ending. This attitude appears in all but two of the eight cases 
where the curve shows a plateau, and it appears in only three 
of the other twelve drawings of the series. This attitude often 
gives way immediately to a reflective attitude of uncertainty 
toward the process as a whole, and this attitude of uncertainty 
seems closely bound up with the visual image which comes up 
just after the completion of the drawing. 

Subject F. — The model used with F. in Series II consisted of 
four slant lines, 27 mm. in length, put together somewhat in the 
form of an M. For Series Ha the model was the same figure 
inverted. The first figure is illustrated in Fig. 4, No. 4, and the 
second one in Fig. 4, No. 5. In both these figures the subject was 
instructed to draw the lines all from top to bottom, this being 
the way that seemed most natural for him. There are certain 
similarities and certain differences in the process of drawing 
these two types of figures. Lines 1 and 3 of the first type cor- 
respond to lines 2 and 4 of the second in that they are drawn 
from upper right to lower left, and are parallel to each other. 
Similarly, lines 2 and 4 of the first figure correspond to lines 1 
and 3 of the second in that they are drawn in the other oblique 
direction and form a parallel pair. The lines of both figures are 
the same in that they are drawn from top to bottom in each case 
and are slanted at the same angle from the perpendicular. An 
important difference between the two figures is that in the first 
the subject is drawing down from points, while in the other he 
is drawing down to points. There is a difference in the mechan- 
ical difficulty of drawing a line from upper right to lower left 
and in the "other oblique direction. The former direction the 
subject pronounced the easier, mentioning it several times in the 
course of the series. 

The method of measurement was as follows. First, the length 



EXPERIMENTAL STUDY OF CONSCIOUS ATTITUDES 207 

of each line was measured in millimeters. Then the straightness 
of the line was indicated by a figure expressing the number of 
slight deviations that occurred in it. The direction of the line 
was determined and expressed in terms of the number of milli- 
meters by which the line deviated from the correct slant. Finally, 
the amount by which the successive lines failed to meet each other 
at the points was also determined in millimeters. 

For time measurement the four lines were divided into halves. 
The time was determined in each line, first for the period at the 
start during which the pencil was down but not moving, then 
for each of the two halves, and then for the period at the end 
during which the pencil was at rest. The time elapsing between 
the end of one line and the beginning of the next, or the time in 
the points, was also determined. 

Table IV gives the accuracy averages for the twenty drawings 

Table IV 

Subject F. 
_ 



No. 


Lgth. 


Str. 


Dir. 


Points 


1. 


30.9 


1.1 


1.6 


572 


2. 


23-9 


1-7 


i-3 


4-75 


3- 


26.5 


1.2 


3-2 


3.25 


4- 


21.7 


2.1 


1.3 





Ha 



No. 


Lgth. 


Str. 


Dir. 


Points. 


1. 

2. 

3- 
4- 


29.0 
26.3 
22.6 
28.1 


2.0 
1.6 

i-5 
1.9 


4-3 
2.3 

2.5 
2.3 


4-05 
3.22 

445 



of each series. It will be seen from this table that in Series II, 
lines 1 and 3 are more accurate, except in direction, than 2 and 4, 
while in Series Ha just the opposite is the case. So, the lines 
which are mechanically easiest to draw are those which are 
drawn most accurately. There is a certain correspondence be- 
tween parallel pairs in the same figure, and in each series line 1 
is drawn longer than any of the others. 

The time records for Series II and Ha are given in Table V — 
in each case the average of twenty experiments. The smaller 
time determinations — those for the parts of the lines — do not 



208 



JOHN TRUMBULL METCALF 



show any marked uniformities and are omitted from this table. 
Under "Time" is given the average time for each line. Under 
"Rate" is given the average rate at which it is drawn. The rate 
is determined by dividing the length of each line by the time 
it took to draw it. This gives the rate in millimeters per second, 
and the average of these determinations appears in the table. 
Under "T. P." is given the average time between the ending of 
one line and the beginning of the next. 

Of the columns in this table those for rate are the most im- 
portant. In II the rate for line i is higher — a greater number 
of millimeters per second — than that of any of the other lines. 
As a pair too, i and 3 are more rapidly drawn than 2 and 4, just 

Table V 

Subject F. 

_ _ 



No. 


Time 


M.V. 


Rate 


M.V. 


T.P. 


1. 


4-27 


.38 


7.40 


.96 


.92 


2. 


4.32 


-37 


5.67 


.87 


1.08 


3- 


4.10 


.46 


6.47 


1.03 


.82 


4. 


4.44 


•43 


4.85 


•57 





Ha 



No. 


Time 


M.V. 


Rate 


M.V. 


T.P. 


1. 


436 


-33 


6.69 


.65 


.84 


2. 


« 3-69 


-36 


7.0S 


•77 


.86 


3. 


3.86 


•33 


5.92 


.56 


.89 


4. 


3-68 


•31 


7.67 


.92 





as in the accuracy table they are found to be drawn longer. In 
Series Ila we find that 2 and 4 are more rapidly drawn than 
1 and 3, though 1 is drawn fairly fast. The accuracy table shows 
that these fast-drawn lines are long, so that in general the ten- 
dency to draw at a fast rate and the tendency to make the line 
long go together. 

From the introspections it is possible to get in a general way the 
order of certainty of the four lines in Series II and Series Ila. 
There is sometimes enough difference between the four lines in- 
trospectively to enable the subject to name them in their order of 
certainty. Where this is not given it is often possible to tell from 
the general character of the introspective report what this order 



EXPERIMENTAL STUDY OF CONSCIOUS ATTITUDES 209 

was. The order of certainty most frequently given in Series II 
was 1, 3, 2, 4. This order was given five times. The order of 
certainty is given in all only nine times, and no other order of 
certainty appears more than once. This order receives support 
also from the introspections in which the certainty of one or two 
lines is spoken of, but the complete order not given. Line 1 is 
called most certain in every case, line 4 is called less certain than 
line 2 in all but two cases. The general outcome of the intro- 
spections of Series II, as far as the order of certainty is con- 
cerned, is that line 1 is the most certain line, that 1 and 3 are 
more certain than 2 and 4, and that the most usual order of cer- 
tainty is 1, 3, 2, 4. 

Recurring for a moment to the accuracy and time records, we 
find that the most certain line 1 is the longest and most rapidly 
drawn. Line 3 is next in length and rapidity, and then follow 
lines 2 and 4 respectively. 

The introspections of Series Ha are somewhat more definite 
than those of Series II. Line 1 is again most certain in all but 
two of the twelve cases in which the order of certainty is given. 
Lines 2 and 4 are much more certain in this series than in the 
other one, and line 3 is called least certain every time it is com- 
pared with the others. The typical order of certainty for this 
series is 1, 4, 2, 3. 

Turning to the records of accuracy and time for this series, 
we find that the certainty of line 1 and of lines 2 and 4 goes with 
greater length of line and faster rate of drawing. The greater 
length and faster rate of line 1 is, however, not as marked here 
as in the other figure where it is drawn in the other easier oblique 
direction. Lines 2 and 4, which are more certain in this figure 
than in the other one, are also drawn at a faster rate here than 
they were there. Line 3, which is least certain in this series, is 
drawn most slowly and is shortest. So it tends to hold here too 
that certainty goes with a long line rapidly drawn. 

One other point must be considered before we go on to the 
pressure records. It was emphasized above that lines drawn 
from upper right to lower left were mechanically easier for the 
subject to execute than the lines drawn in the other oblique 



210 JOHN TRUMBULL MET CALF 

direction. Does certainty simply go with this mechanical ease of 
drawing? In Series II it evidently does. Lines i and 3 are 
more certain than the mechanically harder lines 2 and 4. If it 
is true of the other series as well, lines 2 and 4 should be more 
certain than lines 1 and 3. We find that they are more certain 
than 3, but not as certain as 1. If we should try to show a 
correlation between certainty and the mechanical ease of draw- 
ing, we would be unable to explain how line 1, mechanically easy 
in one case and mechanically difficult in another, is yet most cer- 
tain in each case. The fact that it is the first line drawn in each 
case doubtless has something to do with it, but this will not help 
our explanation unless the fact that it is the first line drawn 
makes the process of drawing it characteristic in some way that 
will show in our records. In the cases of the other lines we have 
found a correlation of certainty with length of line and fast rate. 
This does not hold in the case of line 1 in Series Ha in compari- 
son with lines 2 and 4 of its own series, for although it is longer 
and drawn at a faster rate than the other three lines which are 
drawn in the same direction (II, 2 and 4, Ha, 3) it is not drawn 
faster than, nor is it much longer than, lines 2 and 4 of its own 
series. Its certainty must therefore be connected with some other 
factor which we have not yet observed. 

We find such a factor the moment we examine the pressure 
curves. The pressure curves for line 1 conform to the same type 
throughout both series, except two cases in Series Ha where 
alone line 1 is called uncertain. The form of the curve is a 
gradual steady rise to a maximum which is held without decrease 
to the end. Fig. 7 shows two curves (Nos. 1 and 2) for line 1, 
the first from Series II, the second from Series Ha. No. 3 is 
a curve from one of the two drawings in Series Ha in which 
line 1 was called uncertain. Its variation in form is marked. We 
find here, therefore, in line 1, that certainty and a definite type 
of curve go together, and that this holds regardless of the 
mechanical ease or difficulty with which the line is drawn. 

In the later lines of the drawings the curves show greater pres- 
sure than in line 1, and a somewhat different shape. They rise 
more quickly and their increase is not quite as uniform. Fig. 7, 



EXPERIMENTAL STUDY OF CONSCIOUS ATTITUDES 211 



H-. 




8. 





Fig. 7. 



No. 4, shows a pressure curve from line 2 in the first figure 
(Series II) in which 2 was called next to 1 the most certain line 
in the drawing. Below it is given in No. 5 an example in which 
this line was called least certain of the four. Fig. 7, Nos. 6 and 
7, show examples of line 3 in Series Ha. In 6 the line was cer- 
tain, in 7 uncertain. In 8 and 9 of Fig. 7 there are illustrated 
two pressure curves of line 4 in the first figure (Series II). In 
No. 8 line 4 was certain, in No. 9 it was uncertain. 

This correlation between certainty and a definite type of curve, 
and uncertainty and a variation from that type holds in every 
case for line 1 in both figures. It also holds where a line, usually 
certain, is pronounced particularly uncertain. But it does not 



212 JOHN TRUMBULL METCALF 

hold between small differences in certainty. There are examples 
in the records of both figures where there is no appreciable dif- 
ference between the curves of lines 2 and 3, and yet one is called 
more certain than the other. Evidently, then, only wide con- 
trasts in subjective certainty show corresponding differences in 
the pressure records under the conditions of this experiment. 

Summarizing the results of this experiment with subject F. 
we find that: — 

(1) In general certainty goes with a longer line and a faster 
rate of drawing than does uncertainty. 

(2) The lines which are mechanically easy for the subject to 
execute are usually the lines drawn in this way. 

(3) Line 1 is most certain in each figure, though in the first 
it is mechanically easy and in the second mechanically difficult. 
This certainty is correlated with a definite form of pressure curve 
which is followed in the drawing of line 1 in both figures, and 
is departed from only in the two cases in which line 1 is called 
uncertain. 

(4) In the other lines marked contrasts in subjective certainty 
are reflected in the pressure curves. 

Subject R. — In Series II the experiment with R. was identical 
with that of the practice series after the preliminary line had 
been added to the other three. It consisted in drawing the four 
parallel vertical lines from top to bottom. In Series Ha a varia- 
tion was introduced. This was simply that instead of being 
drawn from top to bottom the lines were drawn from bottom 
to top. 

One change was made in the method of measurement. Instead 
of dividing the lines into thirds and finding the time for each 
third, the lines were divided into halves. Twenty drawings were 
made in each series, and the usual full introspections were taken. 
The subject was able in almost every case to name the lines in 
their order of certainty. No uniform order of certainty de- 
veloped. The succession of drawing, or the position of a line 
in a drawing had no effect upon its certainty or uncertainty. 

A comparison of the accuracy and time averages of Series II 
and Series Ha shows that the lines when drawn from bottom to 



EXPERIMENTAL STUDY OF CONSCIOUS ATTITUDES 213 

top (Ha) are longer, less straight (though as good in general 
direction), and more slowly drawn, than the lines of the other 
series where they are drawn from top to bottom. These differ- 
ences cannot be correlated with a general difference in certainty 
between the two sets of lines. The introspections reveal no gen- 
eral difference which would enable us to say that one series had 
been more or less certain than the other. Yet, since the two 
series show such marked differences in the accuracy and time 
records, we must keep the two series separate when we compare 
the records of the most certain lines with those of the least cer- 
tain ones, as we now proceed to do. 

The averages of the most certain lines and least certain lines 
are given in Table VI. The smaller time determinations are 
omitted as they show no differences between the most certain 
and least certain lines not expressed in the time for the whole 
line. The average rate, however, is included in the table. 

Table VI 
Subject R. 



II 


Lgth. 


M.V. 


Straightness 


Time 


M.V. 


Rate 1 

24.2 


M.V. 


Most 


254 


3-1 


(.9) 1. 


1.07 


.22 


4o 


Least 


24.6 


2.8 


(-94) 1.1 


I.IO 


.22 


23.0 


3-2 


Ila 
















Most 


31.9 


2.2 


(1.8) 1.3 


1.46 


.21 


22.2 


2.2 


Least 


32.6 


1.8 


(1.8) 1. 


1-57 


.24 


21.7 


3. 



In accuracy the most and least certain lines are just about the 
same. In the time values, however, there is a difference, the 
most certain lines being drawn at a slightly faster rate than the 
least certain whether the lines are made from top to bottom or 
from bottom to top. 

As has been said, the subject gave the order of certainty of 
the lines in every case. This order of certainty was, therefore, 
compared in every case with the characteristics of the pressure 
curves. It soon appeared that the line which came first in the 
order of certainty always had a definite form of pressure curve, 
and that the other lines differed from this type in a greater or 
less degree according to their degree of certainty. So marked 
was this form of curve and so universally did it hold, that it was 
possible to tell solely from an examination of the pressure curves 



214 



JOHN TRUMBULL METCALF 





00 

6 

i— i 



oo 



/ 



«/> 



IT) 



EXPERIMENTAL STUDY OF CONSCIOUS ATTITUDES 215 

which line had been called most certain, and in nearly every case 
the order in which the other two followed it. This was true 
both of Series II and Series Ha. The "certainty" form of curve 
is the same for the drawings of each series. Fig. 8 shows some 
examples of curves taken from both series. Nos. 1 and 2 are 
certainty curves from Series II, Nos. 3 and 4 certainty curves 
from Series Ha. It will be observed that the curves follow the 
same form whether the lines are drawn from top to bottom or 
from bottom to top, reversing the mechanical process of drawing. 
The remaining curves in Fig. 8 are tracings of uncertainty curves 
from both series. They vary from the certainty type in two 
ways. Either the increase at the start is too rapid, as in No. 5, 
or the increase is more gradual as in No. 6. No. 7 is a curve 
from a line which the subject called "uncertain at the end." This 
curve shows a decrease of pressure at the end. 

Not only does the curve for certainty follow a definite form, 
but it shows a regularity in its height as well. In the great ma- 
jority (80%) of the cases the pressure curve for the most certain 
line is between the curves for the other two lines in height. So 
there is a general tendency for the certain line to be that one 
which is drawn with a medium pressure. 

This experiment, therefore, has shown certainty in a certain 
part of the drawing process going with a fast rate of drawing 
in that part and with a definite form of reaction as shown by 
the pressure curve. The accuracy results which show no differ- 
ence in accuracy between the certain and uncertain drawings are 
interesting as showing that an accurate drawing does not neces- 
sarily represent the kind of reaction characterized by certainty. 
The accurate drawings vary as to certainty or uncertainty with 
the character of their pressure curves. The same is true of the 
inaccurate ones. 

Series III 
In this series no model was used and the subjects were all 
given the same task. They were instructed to make a dot, raise 
the pencil and move it to a point perpendicularly above the dot, 
and draw a straight line down to, and ending in, the dot. The 
distance above the dot to which the pencil should be moved was 



2i6 JOHN TRUMBULL METCALF 

not designated. The records of the process included everything 
from the making of the dot to the end of the downward stroke. 
After a series of ten experiments had been made in this way, the 
subject was given special instructions each time as to the rate 
at which he should draw the downward line. Sometimes he was 
told to draw it slowly, at other times rapidly, and at other times 
at his natural speed. The instruction changed from experiment 
to experiment. This enabled the subject to get an introspective 
comparison between the experiences of drawing at different rates. 
About ten experiments were made with each of these special in- 
structions, so that in this series there were in all about forty 
experiments with each subject. 

The measurements for accuracy included the length of the line 
in millimeters, its straightness, and the amount by which it varied 
from the vertical where there was such variation. Where the 
line failed to strike the dot the horizontal and vertical distances' 
from the end of the line to the dot were recorded in millimeters. 
The time measurements included the time consumed in making 
the dot, the time between the making of the dot and the beginning 
of the downward line, and the time for the downward line with 
the usual minor determinations. 

The purpose of this arrangement is to provide a drawing pro- 
cess which shall be as simple as possible. The downward line is 
now drawn with but one determining purpose — to reach the dot. 
The experiments with the different subjects will be considered 
separately. 

Subject A. — A. reports a good deal of visual imagery. Atti- 
tudes of certainty and clear visual imagery go together. This 
fact is observed and commented upon by the subject. He said 
once when he gave his introspections for an uncertain drawing, 
"I always feel when a line is uncertain that the visual scheme 
is weak." 

The tables in this series give the number of attitudes of cer- 
tainty and uncertainty reported by the subject in each one of the 
four parts of the series. They also give the average length, time, 
and rate of the lines for the cases of certainty and uncertainty 
respectively. 



EXPERIMENTAL STUDY OF CONSCIOUS ATTITUDES 217 

Table VII 
Subject A 

No special instruction 



Attitude 


No. Length Time 


Rate 


Certainty 
Uncertainty 


4. 2i.i 3-33 
6. 23. 3.19 

Special instruction: — Natural speed 


6.47 

7-35 


Attitude 


No. Length Time 


Rate 


Certainty 
Uncertainty 


3- 18.5 2.49 
6. 20.7 2.60 

Special instruction : — Slow 


7-37 
8.71 


Attitude 


No. Length Time 


Rate 


Certainty 
Uncertainty 


4. 18.9 4.20 

5. 23.4 5-13 

Special instruction : — Fast 


4.46 
4.70 


Attitude 


No. Length Time 


Rate 


Certainty 
Uncertainty 
No attitude 


2. 16.5 1.21 
1. 17. 1.30 
3- 16.6 .99 


13-6 

13. 

18.4 



This table shows no marked correspondence between attitude 
and length, or between attitude and time. If one rate — either 
fast or slow — were especially favorable for the appearance of 
certainty we should expect a larger percentage of attitudes of 
certainty in the experiments in which the instruction to draw at 
that rate was given. In general the certain drawings are drawn 
at a somewhat slower rate than the uncertain ones. The special 
instructions do not greatly affect the relative number of attitudes 
of certainty and uncertainty which appear. In the fast series 
when the rate becomes very fast no attitude appears. 

As for the accuracy with which the subject carried out his task, 
the measure for this is given in the average horizontal and verti- 
cal errors. The horizontal error is the distance from the dot to 
the downward line, and the vertical error the vertical distance 
between the end of the line and the level of the dot. The average 
errors for all cases of certainty and all cases of uncertainty are 
as follows: 

Horizontal Vertical 

Certainty .8 1.9 

Uncertainty .5 3.3 

This shows that in cases of certainty, on the average, the end 
of the line comes nearer to the level of the dot. There is only 



218 



JOHN TRUMBULL MET CALF 



a slight difference in the horizontal error, it being slightly greater 
for the certain lines. 

The pressure records in this series for this subject are very 
unsatisfactory, because he drew with such very light pressure 
that in many cases the rise in the curve was barely perceptible. 
No correlation is possible, therefore, between certainty and a 
definite type or amount of pressure. One experiment in this 
series, however, is especially interesting. In one drawing the 
subject was instructed to draw at the rate which would give him 
most certainty and "exert a little more pressure." The result 
was that the subject, following these instructions, drew with 
great certainty. He pronounced it "as good certainty as I've 
had." 

In general, however, we cannot draw any conclusions from 
this series with this subject except the one derived from the 
introspections, that an attitude of certainty is always accom- 
panied by a clear visual image. 

Subject C. — With C. the introspections are quite different in 
character. He relies mainly upon kinaesthetic factors to guide 
him in the drawing process. Table VIII follows the plan of the 
preceding table. 

Table VHI 
Subject C 

No special instructions 



Attitude 


No. Length 


Time 


Rate 


Certainty 
Uncertainty 


10 18.9 
— 


2.60 


7.o 




Special instruction: — Natural speed 




Attitude 


No. Length 


Time 


Rate 


Certainty 
Uncertainty 
No attitude 


6 16.8 

— 

1 18. 


2.03 
1.52 


8.3 
11.8 




Special instruction :- 


-Slow 




Attitude 


No. Length 


Time 


Rate 


Certainty 
Uncertainty 


2 13-5 

8 15.8 


4.04 
6.46 


3-33 

2.52 




Special instruction :- 


—Fast 




Attitude 


No. Length 


Time 


Rate 


Certainty 
Uncertainty 


9 19. 

1 17. 


1.13 

.65 


18.6 
26.1 



EXPERIMENTAL STUDY OF CONSCIOUS ATTITUDES 219 

It is apparent at once from these tables that with this subject 
the rate of drawing has a decided influence upon the attitudes 
of certainty. His normal rate, as shown by the drawings made 
with no special instructions, is accompanied by certainty in every 
case. In the experiments with special instructions the ones in. 
which he is instructed to draw at natural rate are faster than 
those in which no special instructions are given. In the experi- 
ments with the instruction to draw slowly the drawing is made 
very slowly, except in two cases in which the subject reported 
that he had forgotten the instruction. Great uncertainty was 
reported throughout these drawings, except in the two cases in 
which the instructions had been forgotten. The instruction to 
draw fast increases certainty. The fast drawings are all called 
particularly certain by the subject — more so even than those 
made with no special instructions. Only one drawing in this 
series shows a trace of uncertainty. This is the fastest of all 
the fast drawings, and we shall have occasion to refer to it again 
when we consider the pressure records. 

The horizontal and vertical errors for the different series are 
as follows: 

Horizontal Vertical 

No special instruction 1.35 1.3 

Instruction — Natural Speed .60 .8 

Instruction — Slow 1.20 3.5 

Instruction — Fast 1.25 .85 

The smaller error in the Natural speed series may be attributed 
to practice, since these were made after the drawings with no 
special instructions. The slow series in which all but one of the 
cases of uncertainty occur, shows a large vertical error, and so 
a correlation between uncertainty and inaccuracy. The inac- 
curacy is due to the fact that the subject stops drawing before he 
has come down to the level of the dot. In his introspections he 
says in one place that "in the slow drawings the length of the 
line seems greater." Evidently, then, when he draws slowly he 
overestimates the length of the line he is drawing. 

The lines drawn with no special instructions were all certain, 
and their pressure curves all conform to a fixed type. This curve 
rises slowly and then shoots up to form a plateau at the end. 



220 



JOHN TRUMBULL METCALF 



This type of curve was found to go with an attitude of certainty 
when the triangle was used as a model and the last stroke was 
a downward one. A typical example is given in Fig. 9, No. 1. 



*. 



3. 



/ 



/ 



v,. 



Fig. 9. 



In the series with the special instructions to draw at natural 
speed, the curve continues to show the certainty type in three 
of the seven cases, but in four cases it is different. The curves 
of these four cases are like each other, however, and one of them 
is illustrated in Fig. 9, No. 2. The introspections do not show 
that the drawings for which these are the records are markedly 
different from the others except that they are somewhat less 



EXPERIMENTAL STUDY OF CONSCIOUS ATTITUDES 221 

certain. In one case certainty was reported. In another there 
was no attitude. In a third, certainty was reported, but it was 
called less in degree than that of the previous experiment, the 
curve of which had a plateau. In the fourth case the subject 
said that the line had run out of true about half-way down, but 
had come back, negative certainty giving way to positive. 

The curves for the slow drawings may be briefly dealt with. 
The pressure is very slight in every case, and the plateau at the 
end is present in only one case. Even here it is so slight as to 
be scarcely noticeable. So we find that the following out of the 
special instruction to draw slowly has affected not only the atti- 
tude of certainty with which the drawing is made, but also the 
amount of pressure exerted and the form of the curve. 

In the fast drawings the typical certainty curve is present in 
all but two cases. The curve in these cases is more like the vary- 
ing form that was observed in the series with the special instruc- 
tion to draw at natural speed. One of these is the drawing 
referred to above as the fastest of all the fast drawings — the 
only case of uncertainty outside the slow series. This pressure 
curve is shown in Fig. 9, No. 3. No. 4 shows a typical curve 
from the fast series. 

So we find with this subject that the attitude he reports varies 
with the rate of drawing, and the type of pressure curve varies 
with the attitude. 

Subject F. — In this series F. reports both types of imagery, 
visual-motor imagery being most prominent. With him the ex- 
periment is the same as with the other subjects and the table 
follows the same plan. 





Table IX 






Subject F. 






No special instructions 




Attitude 


No. Length Time 


Rate 


Certainty 
Uncertainty 


6 15-4 3-31 
4 14-5 3.14 

Special instruction : — Natural speed 


' 4-98 
4.80 


Attitude 


No. Length Time 


Rate 


Certainty 
Uncertainty 


4 20.2 2.83 
2 18.7 3.66 


7.26 
5-II 



222 JOHN TRUMBULL METCALF 





Special 


instruction :- 


-Slow 




Attitude 


No. 


Length 


Time 


Rate 


Certainty 
'Uncertainty 
No attitude 


5 
3 

2 


20 
21 
20 


5-51 
531 

5.25 


375 
4.04 

3.85 




Special 


instruction :- 


—Fast 




Attitude 


No. 


Length 


Time 


Rate 


Certainty 
Uncertainty 


8 

2 


2Q 

19 


1.29 
1.64 


16.3 
14.5 



These tables show that the subject was not as much affected 
by the special instructions as was C. The instruction to draw 
slowly does not result in any marked change in the degree of 
certainty or in the number of times it is reported. The instruc- 
tion to draw fast, however, has such an effect, which is to make 
the series in general more certain. The two uncertain drawings 
were the first of the series. The subject had a little trouble in 
getting accustomed to drawing the line fast, and this is reflected 
in the uncertainty of the first two drawings. After these, how- 
ever, the series is composed entirely of certain drawings, and 
the degree of certainty is great. 

With this subject the greater average errors are in the certain 
drawings, so certainty and accuracy in this case do not go to- 
gether. The average errors are as follows : 

Horizontal Vertical 

Certainty .8 2.9 

Uncertainty .65 2.3 

The pressure records with this subject show little difference 
between the certain and uncertain drawings when there are no 
special instructions, or when the instruction is to draw at natural 
speed. On the whole the curves of the certain drawings are more 
regular than those of the uncertain ones, but the difference is not 
great enough to show a definite type of curve for certainty, from 
which the uncertain curves vary. Six of the ten certain draw- 
ings show curves with characteristics similar enough to warrant 
our calling them all of the same form, but three of the six curves 
from uncertain drawings follow this form as well. Fig. 10, 
No. 1, illustrates a curve of this type from a certain drawing. 
When we come to the special instructions "slow" and "fast," 
however, we find that the certain and uncertain drawings have 



EXPERIMENTAL STUDY OF CONSCIOUS ATTITUDES 



223 



different pressure curves. The curves for the certain drawings 
conform to the type found in the previous drawings, while the 
uncertain ones vary from it in some way. In the slow series the 
curves for the two drawings in which definite uncertainty was 
reported are very flat, showing very light pressure and very little 
difference in pressure between the beginning and the end of the 
line. They are less in amount than any of the certain drawings 
except one in which "only fair certainty" is reported. Fig. 10 




X. 




3. 



5. 




Fig. 10. 



224 



JOHN TRUMBULL MET CALF 



gives in Nos. 2, 3, and 4 respectively, tracings from a certain, 
an only moderately certain, and an uncertain drawing. 

In the drawings with the instruction "fast," certainty was re- 
ported in all but the first two cases, in which the subject was 
becoming accustomed to the process. All the curves of this 
series except the first two conform to the type already found to 
accompany certainty. One of them is illustrated in Fig. 10, 
No. 5. 

With this subject, therefore, we have not been able to correlate 
the attitudes of certainty and uncertainty with a definite form 
of curve when the subject has chosen his own rate of drawing, 
or has been instructed to draw at natural speed. When, how- 
ever, he has been instructed to draw slowly or at a fast rate, 
differences in the objective records appear, and these vary with 
differences in attitude. 

Subject R. — R. differs from the other subjects in the amount 
and kind of imagery he reports. His imagery is chiefly visual, 
though kinaesthetic imagery sometimes appears. Often incom- 
plete images are reported, and sometimes part of an image is 
vague while the rest is clear. 



Table X 

Subject R. 

No special instructions 



Attitude 


No. Length 


Time 


Rate 


Certainty 
Uncertainty 


7 16.4 
3 18.8 


1.98 

2.55 


8.9 
7-9 




Special instruction: — Natural speed 




Attitude 


No. Length 


Time 


Rate 


Certainty 
Uncertainty 


4 17.6 
6 15.8 


2.20 
2.39 


8.5 
6.8 




Special instruction :- 


-Slow 




Attitude 


No. Length 


Time 


Rate 


Certainty 
Uncertainty 


3 15.6 
6 15-3 


2.72 
3-97 


5-86 
4.15 




Special instruction :- 


-Fast 




Attitude 


No. Length 


Time 


Rate 


Certainty 
Uncertainty 


7 16. 
3 13.3 


1. 00 
1.20 


15.5 
1 1.9 



EXPERIMENTAL STUDY OF CONSCIOUS ATTITUDES 225 

These tables show a correlation between the rate of drawing 
and the character of the attitudes reported. The averages in all 
four parts of the table show that the certain lines were drawn at 
a faster rate, and (except with "no special instructions") they 
are made longer than the uncertain. 

The accuracy with which the subject carries out his task seems 
not to have any effect upon his certainty. A comparison between 
the average horizontal and vertical errors of the certain and un- 
certain cases is given in the following table. 

Horizontal Vertical 

Certainty 1.1 2.3 

Uncertainty 1.8 1.5 

Turning now to the pressure curves of this series, we find that 
in the experiments with no special instruction, and in the ones 
with the instruction to draw at natural speed, the certain draw- 
ings follow a form which is rather like the form of the curve 
for the certain lines in the experiment with this subject in Series 
II and Ha. The curves for the uncertain drawings show a 
smaller amount of pressure, and they vary from the certainty 
form in other ways. They are little higher at the end than at 
the beginning, and sometimes show a decrease of pressure within 
the drawing of the line. Of the eleven cases of certainty all but 
four have curves of the certainty form. Among the nine cases 
of uncertainty only one shows a curve of the certainty form. 
Fig. 11 shows in No. 1 a curve from a certain drawing and in 
No. 2 one from an uncertain drawing. 

In the slow drawings there are three cases of certainty and 
six of uncertainty. Two of the cases of certainty have curves of 
the certainty form, the curve for the third is flat. The cases 
of uncertainty all have flat curves that show no trace of the cer- 
tainty form. One of these drawings is peculiar, however, in 
that it is of the usual flat uncertainty type until the very end. 
Then there is a little sudden increase in pressure. The curve 
is illustrated in Fig. 11, No. 3. The introspections on this line 
show that during the drawing it was attended by an attitude of 
uncertainty, but that at the very end there appeared an attitude 
of certainty. 



226 



JOHN TRUMBULL METCALF 



6 



*~i 



CN< 



CO 



EXPERIMENTAL STUDY OF CONSCIOUS ATTITUDES 227 

The curves for the fast series are all of the certainty type 
except four. Two of these are curves for drawings which were 
called certain in spite of this variation. An example of a cer- 
tainty curve from a fast drawing is shown in Fig. 11, No. 4. 

With this subject, therefore, certainty tends again to go with 
a long line drawn at a fast rate, and a definite type of pressure 
curve. Uncertainty is correlated, though not perfectly, with a 
shorter line drawn at a slower rate, and a variation from the 
type of pressure curve found in the certain drawings. The ac- 
curacy with which the subject carries out his task, as measured 
by the horizontal and vertical errors, seems to have no effect upon 
the attitude of certainty. 

Summarizing the results for Series III for all subjects, we 
find that there are large individual differences. These appear 
both in the introspective reports and in the objective records. 
With A. clear visual imagery and certainty go together. He is 
much less influenced as to his attitude by varying rates of draw- 
ing than any of the other subjects. Since his pressure records 
were unsatisfactory it is impossible to compare them with those 
of the other subjects, but it may be remarked that here as else- 
where he draws with lighter pressure than any of them. C. relies 
almost entirely upon kinaesthetic factors, and with him rate of 
drawing has a great influence upon attitude, a slow rate going 
with uncertainty and a fast rate with certainty. Moreover, with 
this subject certainty goes with a definite type of pressure curve, 
and when the attitude is modified with change of rate, this form 
of curve is modified as well. F.'s imagery is more visual than 
that of any of the other subjects except A. The correlation be- 
tween attitude and rate is slightly better with him than with A., 
but not as good as with the other two subjects. With F. it is 
not possible to correlate the attitude of certainty with a definite 
form of curve when he has no special instructions or is instructed 
to draw at natural speed. When, however, he is drawing at a 
rate faster or slower than the normal, the attitude of certainty 
is found to be correlated with a definite type of curve. R. re- 
ports less imagery than any of the other subjects. He relies more 
upon kinaesthetic sensations in drawing than any of the others 



228 JOHN TRUMBULL METCALF 

except C. With him there is found a good correlation between 
length, time, rate, and attitude. The special instructions have 
more effect upon the relative number of attitudes of certainty 
and uncertainty reported than in the cases of A. and F., but less 
effect than in the case of C. With this subject (R.) there is also 
a fairly good correlation between certainty and a definite type 
of pressure curve and between uncertainty and some variation 
from this type. 

It is worthy of special comment that the subjects form a series 
with regard to the completeness of the correlation between atti- 
tude and objective records. They also form a series in the 
imagery they report, and they are ranked in the same order in 
each series. With the most visual subject the correlation is least 
complete, and with the least visual and most kinaesthetic subject 
the correlation is most complete. It seems, then, that with a vis- 
ualist the attitude of certainty appearing in the course of a draw- 
ing is less influenced by the rate of drawing and the pressure 
exerted than is the attitude of a subject who relies chiefly upon 
kinaesthetic factors. With the visualist the important thing for 
an attitude of certainty seems to be a clear visual image, with 
little or no relevant change in rate and pressure appearing. With 
the kinaesthetic subject visual imagery has little to do with the 
attitude which, however, does vary with the rate and pressure 
characteristics. 

Series IV and IVa 

This series, in which all the subjects took part, had for its task 
the drawing of a simple figure with the left hand. First the 
drawing was made by the left hand alone (Series IV) and then 
by the left hand with the right hand drawing symmetrically with 
it (Series IVa). In this second series the subject took two 
pencils and made two symmetrical drawings, one with each hand. 
The figures used are illustrated in Fig. 4, Nos. 6 and 7. The 
dotted lines in these figures indicate the path followed by the 
right hand when it drew with the left, though these lines did not 
appear in the model. The same model, illustrating the drawing 
to be made by the left hand, was used both when the left hand 
drew alone and when the right hand drew with it. It was thought 



EXPERIMENTAL STUDY OF CONSCIOUS ATTITUDES 229 

that these conditions would furnish a good contrast between the 
attitudes which appeared when the left hand drew alone and 
when the right hand drew with it. As all the subjects were right- 
handed, drawing even of the simplest figures with the left hand 
would be a very unaccustomed activity. On the basis of previous 
experiments it was thought that the process of carrying out this 
activity, with left hand alone, would be attended by attitudes of 
uncertainty. It was thought too that when the right hand drew 
a symmetrical figure at the same time, the better motor control 
thus gained for the left would result in greater certainty. The 
good motor control over symmetrical movements is well known. 

Accordingly, a series of the drawings from each of two models 
was first made with the left hand alone. This series was com- 
pleted before the series of symmetrical drawings was begun to 
avoid the practice effect which the symmetrical drawings would 
have had upon the others if they had been intermingled with 
them. Then the symmetrical drawings from each of the two 
models were made. The figures were always made continuously, 
that is without raising the pencil from the paper. They were 
begun at the top, and the left hand went first to the left and then 
downward, or, in the case of the second figure, obliquely to the 
right. In the symmetrically drawn figures the right hand fol- 
lowed a corresponding inverse course. 

In this series the usual accuracy, time, and pressure records 
were taken. In the symmetrical drawings the time and pressure 
records were taken of the left-hand drawing only, but the right- 
hand drawing was measured for its accuracy. The symmetrical 
drawings necessitated a special arrangement of the apparatus. 
The right-hand drawing had to be made upon a smooth surface 
exactly beside the left-hand one which was made over the pres- 
sure plate. To provide an arrangement which would satisfy these 
conditions half the opening in the plate, H, Fig. 1, was filled up. 
A piece of zinc the exact size and shape of the right-hand half of 
the opening was cut out and soldered in place, so that the zinc 
simply formed a continuation of the original plate. For the 
right-hand drawings rectangular pieces of paper were cut out 
a convenient size and fixed upon this new portion of the plate. 



230 JOHN TRUMBULL METCALF 

In reporting the results of these experiments the model con- 
sisting of a horizontal and a vertical line, Fig. 4, No. 6, will be 
spoken of as the first model, and the one composed of the two 
slant lines, No. 7, will be called the second model. The lines of 
the first model are shorter than those of the second. In the first 
they are 15 mm. long, and in the second 20 mm. The angle is 
a right angle in each case. 

The measurements of accuracy with the first figure included 
the lengths of the two lines in millimeters, the usual two determi- 
nations of their straightness, and the size of the angle in degrees. 
In the time records the time for each line was measured and the 
time for the drawing of the whole figure determined. In the 
second figure the method of measurement was a little different. 
The length of the two lines was measured in millimeters. There 
was only one determination of straightness, however. This was 
the number of slight fluctuations — the general deviation was not 
measured. The amount by which the two lines vary in direction 
with regard to each other is contained, however, in the measure- 
ment of the angle. The time measurements include the time for 
each line and the time the pencil is at rest at the apex of the 
angle between the two lines. 

Table XI compares the accuracy records of the two series with 
the first model. Opposite "IV" are given the averages for the 
left-hand drawings when they were made with the left hand 
alone. Opposite "IVa" are given the averages for the left-hand 
drawings, and opposite "IVa R. H." the averages for the right- 
hand drawings, in the symmetrical series. The lengths of the 
horizontal and vertical lines are given under "L'gth H" and 
"L'gth V." Their straightness error in terms of the number of 
little deviations under "Str. H" and "Str. V," and their direction 
or amount of total deviation under "D H" and "D V." Under 
"angle" is given the average size of the angle in degrees. The 
figures in parentheses are mean variations. 



EXPERIMENTAL STUDY OF CONSCIOUS ATTITUDES 231 



Table XI 

Accuracy (First model) 

Subject A 





L'gth.H. 


L'gth.V. 


Str.H. 


Str.V. 


D.H. 


D.V. 


Angle 


IV 

IVa 

IVaRH 


12.2 (2. ) 
11.4(1. ) 
10. (1. ) 


11. (i.5) 2.5 
11.2 ( .5) 2.3 

12. (1. ) 1.3 

Subject C. 


2.7 

3. 
14 


1. 
1.8 
.8 


2.2 

3- 
1.4 


105(3. ) 
112(6. ) 
102(3.4) 




L'gth.H. 


L'gth.V. 


Str.H. 


Str.V. 


D.H. 


D.V. 


Angle 


IV 

IVa 

IVaRH 


18.9 (2.3) 
16.5 (2.7) 
18.5 (2.2) 


17.4 (2.6) 1.4 
22. (3. ) 2.4 
18.3 (1.4) i-3 

Subject F. 


3.2 
2.9 
2. 


2. 
1.8 

24 


2.5 
3-9 
2.4 


90(8. ) 

101(9. ) 

90(1. ) 




L'gth.H. 


L'gth.V. 


Str.H. 


Str.V. 


D.H. 


D.V. 


Angle 


IV 

IVa 

IVaRH 


20.8 (1.6) 
15.8 (2.7) 
14.1 (1.9) 


18.6 (2.1) 
15.3 (2.9) 
15.5 (1.8) 

Sub 


34 

3- 

1. 

ject R. 


2.5 
2.4 

2. 


2.6 

2. 

1. 


2.3 
3-3 

1. 


94(6. ) 
107 (6.6) 

90(3.7) 




L'gth.H. 


L'gth.V. [Str.H.2 


Str.V. 


D.H. 


D.V. 


Angle 


IV 

IVa 

IVaRH 


16.3 (2.5) 

17.5 (2.9) 
18. (2. ) 


15.2 (1.6) 
17.5 (3-6) 
20.8 (2.9) 


2. 

1.9 
1.9 


2.7 

3-1 
2.2 


1-7 
4.6 
2.8 


2. 

2.9 

4. 


9i(4.5) 
85(7-5) 
00(3. ) 



This table shows that the addition of the right hand in the 
process improves slightly the accuracy with which the subject 
makes the drawing with his left hand as far as the equality of 
the lines is concerned, except in the case of C. With all the 
subjects, however, the errors in straightness, direction, and angle 
are greater when the right hand draws with the left than when 
the left hand draws alone. The drawing made with the right 
hand is in general more accurate than either of the drawings 
made with the left. 

Table XII gives the accuracy records for the second model. 
Under "L'gth. 1" and "L'gth. 2" are given the average lengths 
of the first and second lines. The straightness values for the two 
lines and the average measurement of the angle follow in the 
other three columns. Here it will be observed that the addition 
of the right hand does not have as much effect upon the accuracy 
of the left-hand drawing as far as the equality in length of the 
lines is concerned. The straightness of the lines, however, is 
less in every case when the right hand is drawing with the left. 
The exactness with which the angle is approximated is not much 
affected. 



'32 



JOHN TRUMBULL METCALF 



IV 

IVa 

IVaRH 



IV 
Va 
IVaRH 



Table XII 

Accuracy (Second model) 

Subject A. 



L' 



gth.i. 



19. (1.8) 
17.3(1.8) 
14-5 (1. ) 



L'gth.2. 



15. (14) 
13.4 (1.2) 

14.7 (2.2) 



Str.i. 



2.8 

3-5 

2. 



Subject C. 



L'gth.i. 



19-3 (3.9) 
18.3(1.9) 
14.9 (1.1) 



L'gth.2. 



23- (3.i) 

20.9 (2.7) 
16.1 (1.8) 



Str.i. 



1.9 
2.7 

2.3 



Subject F. 



Str.2. 



2.6 
2.2 
1. 



Str.2. 



1.6 
1.2 
1-3 



Angle 



94.6(4.8) 

94- (7- ) 
101. (64) 



Angle 



88. (6.9) 
90. (5- ) 
95. (4. ) 



IV 

IVa 

IVaRH 



L'gth.i, 



27.6 (3. ) 
24.4 (2.2) 

19-7 (34) 



L'gth.2. 



23.6 (2.8) 
22.4 (2.4) 

15. (2.5) 



Str.i. 



2.5 
5-1 
1-7 



Str.2. 



1-3 
44 
1.7 



Angle 



94- (6.8) 

84.(6. ) 

103.(6. ) 







Subject R. 








L'gth.i. 


L'gth.2. 


Str.i. 


'S'tr.2. 


Angle 


IV 
Va 
IVaRH 


21. (1.8) 
244 (2.8) 

22. (1.2) 


20.6 (1.9) 
22.6 (3.5) 
21.9 (2.3) 


2.6 

3-9 
2.8 


3-3 
3-5 
1.9 


100.(10.) 
82.(9.6) 
93. (6-6) 



The averages for the time records of the drawings made from 
the first model are given in Table XIII. This table gives under 
"Horiz." the time for the horizontal line, under "Vert." the time 
for the vertical line, and under "Total" the time for the whole 
drawing. The figures in parentheses are mean variations. 





Table XIII 

Time (First model) 

Subject A. 






Horiz. 


Vert. 


i ai 


IV 
IVa 


346(.33) 
3.60O25) 

Subje 


3-02 (.52) 
3.26(43) 

:ct C. 


648(72) 
6.86(49) 




Horiz. 


Vert. 


Total 


IV 
IVa 


3-99 (-74) 
3.92(.49) 

Subje 


346 (.66) 
2.60(48) 

;ct F. 


7.37(1.35) 
6.52(1.08) 




Horiz. 


Vert. 


Total 


IV 

IVa 


4.02 (.65) 
4-44 (.68) 


3.52(43) 
3.8i (.56) 


7.54(1.07) 
8.25(1.13) 



EXPERIMENTAL STUDY OF CONSCIOUS ATTITUDES 233 





Subje 


:ct R. 






Horiz. 


Vert. 


Total 


IV 
IVa 


2.42 (.30) 
3.i4(.2S) 


2.03 (.27) 
2.79 (-37) 


4-45 (.52) 
5-93 (.59) 



This table shows that with all the subjects except C. the addi- 
tion of the right hand results in slower time. With C. it results 
in a faster time. The time records for the second model are 



given 


in Table XIV. 


Table XIV 








Time 


(iSecond mo 
Subject A. 


del) 








1. 


Point 


2. 


Total 




IV 
IVa 


3.26 

3-55 


•50 
.49 


2.61 

2.77 


6.37 
6.81 






Subject C. 










1. 


Point 


2. 


Total 




IV 
IVa 


367 

4.00 


1-47 
1. 12 


3-34 
3.25 


8.48 
8.37 






Subject F. 










1. 


Point 


2. 


Total 




IV 
IVa 


3.81 
4.82 


•95 
.69 


3.64 
4-54 


8.40 
10.05 






Subject R. 










1 

1. 


Point 


2. 


Total 




IV 
IVa 


2.22 
2.78 


•59 
•49 


2.29 
2.42 


5.10 
5.69 



This table shows that the addition of the right hand gives a 
slower rate of drawing for all subjects except C. With him the 
rate is slightly faster again. 

The introspections, in so far as they have to do with the rela- 
tive certainty with which the figures are drawn under the two 
different conditions, give a wholly unexpected result. It appears 
from them that the addition of the right hand does not increase 
the subject's certainty with regard to the left-hand drawing, but 
decreases it instead. The introspections, however, furnish the 
explanation for this result. All the subjects reported that their 
attention was for the most part upon the left-hand drawing, but 
that it would shift over at times to the right-hand one. This 



234 JOHN TRUMBULL METCALF 

would occur several times in the course of the drawing of a given 
figure. It occurred most often in the case of F., who reported 
that attention oscillated between the two drawings as many as 
seven or eight times in the course of the process. This shifting 
of the attention occurred least often in the case of C. He was 
the only subject who seemed able to keep his attention upon both 
drawings at once. The time results show that C. was the only 
subject for whom the addition of the right hand did not result 
in longer drawing times. 

The pressure records may be briefly dealt with. They do not 
show a different form of curve for each series. The typical form 
of the record of pressure changes is the same for the subject's 
left hand whether it is drawing alone or with the right hand 
drawing at the same time. Thus we have two series, one more 
certain than the other, yielding pressure curves which do not re- 
flect this difference in attitude. The accuracy records of the two 
series are little different. The time records do show a difference 
between the series, for in the case of all the subjects except C. 
the time is somewhat longer for the less certain series. This, 
however, by itself is not very significant, and for the explanation 
of these results we must turn to the introspections already men- 
tioned. The addition of the right hand in the drawing process, 
changing as it does the subject's total reaction, brings about a 
change in attitude too, and this change is experienced intro- 
spectively as a dividing of the attention, or a shift of the atten- 
tion from the left hand. Thus, although the objective form of 
the left hand's reaction remains the same (except in its time 
characteristics), the change in the subject's total reaction has 
brought about a change in attitude. Introspectively this is desig- 
nated as distraction of the attention, and it is natural that a 
slower rate of drawing should be found to accompany it. The 
case of C. is worthy of special comment. Of the four subjects 
he was least affected with regard to his attitude of certainty by 
the addition of the right hand. He is also the subject who has 
shown himself throughout to depend most upon kinaesthetic sen- 
sations and images. He seems to have had less difficulty than 
any of the others in adopting the new form of symmetrical draw- 



EXPERIMENTAL STUDY OF CONSCIOUS ATTITUDES 235 

ing. His time records show that instead of lengthening the time 
of the drawing, as is the case with the other subjects, the addition 
of the right hand actually shortened it a little. It appears, then, 
that the subject who makes greatest use of kinaesthetic factors 
is least affected by the change in the form of the total reaction 
made by the addition of the right hand in the drawing process. 

Summary and Conclusion 

The purpose of our investigation was to see whether there was 
a correlation between the conscious attitudes experienced by the 
subject in connection with his reaction to certain instructions, 
and the objective form of the reaction as measured by its ac- 
curacy, rate, and pressure.* We now summarize briefly the re- 
sults of the experiments. 

In the first place, the practice series yields nothing of final 
value except the indication that the attitude of certainty in some 
given part of the process may be correlated with a slightly faster 
rate of drawing. 

Series II, Ha and lib, in which each subject was given a dif- 
ferent task, yield results which differ in their definiteness, some 
of the figures being better suited to the purposes of the experi- 
ment than others. With A., who drew the two series of circles, 
the first anti-clockwise and the second clockwise, the introspec- 
tions showed that the second halves of the circles were always 
less certain than the first halves. The time records show that 
the rate of drawing increases during the first half and decreases 
during the second half. The pressure records show that in both 
series the pressure of the first half rises gradually and regularly 
with never a decrease, while in the second halves this regular in- 
crease is not continued, irregularities occur, and sometimes a de- 
crease in pressure appears. In the anti-clockwise circles a point 
of maximum uncertainty was located in six of the experiments. 
This point always came in the third, or between the third and 
fourth quarters of the circle. The time records show that when 
it is thus located it comes at the point at which the greatest re- 
tardation in rate is taking place, and the pressure records show 
that at this point too there is a decrease in the amount of pres- 



236 JOHN TRUMBULL METCALF 

sure. Thus, in this series for this subject certainty seems to be 
correlated with an increasing rate of drawing and steadily in- 
creasing pressure ; uncertainty with a decreasing rate and irregu- 
lar pressure characteristics. 

The triangle figure used in Series II and Ha with C. proved 
to be very unsatisfactory. The results did, however, give some 
indication that the attitude of certainty is accompanied by a more 
regular pressure curve than the attitude of uncertainty. Where 
the figure ended in a downward stroke (Series Ha) an attitude 
of certainty was sometimes reported during the drawing of the 
last line. When so reported the pressure curve of the drawing 
was found to end in a "plateau." 

With F., whose task in Series II and Ha was the drawing of 
four slant lines combined in two different ways, it is found that 
in general certainty goes with a longer line and a faster rate of 
drawing than does uncertainty, and the lines which are mechan- 
ically easy for the subject to execute are usually the lines drawn 
in this way. This, however, does not necessarily hold through- 
out, because the first line in each drawing is the most certain, 
although in the first drawing it is mechanically easy and in the 
second mechanically difficult. This certainty is correlated with 
a definite form of pressure curve which is followed in the draw- 
ing of the first line throughout, and is departed from only in the 
two cases in which this first line is called uncertain. In the other 
lines marked contrasts in subjective certainty are reflected in 
differences in the pressure curves, but small differences are not 
so reflected. 

The most satisfactory figure in Series II and Ha proved to 
be the four vertical lines which formed the task for R. He was 
usually able in his introspective report to rank the lines in their 
order of certainty, and this gave a good basis of comparison 
with the records of accuracy, time, and pressure. The most cer- 
tain and least certain lines of each experiment were compared 
for accuracy and time. It was found that there is practically no 
difference in accuracy between the most certain and the least 
certain' lines, but the most certain lines are drawn at a somewhat 
faster rate than the least certain. As for pressure, it is found 



EXPERIMENTAL STUDY OF CONSCIOUS ATTITUDES 237 

that there is a very definite form of curve which always goes 
with certainty, and a variation from that form of curve always 
goes with uncertainty, this variation being greater or less accord- 
ing as the degree of uncertainty is greater or less. In most cases 
also the curve for the most certain line is between the other two 
in height, this holding in 80% of the cases. These results hold 
both in Series II, in which the lines were drawn downward, and 
in Series Ha, in which they were drawn upward. Thus, in this 
series with this subject we find that the accuracy of the drawing 
has no effect upon the attitude experienced in connection with it. 
The time, however, is less for the certain than for the uncertain 
lines, and a definite form of pressure curve is found to go with 
certainty, and some variation from this form with uncertainty. 
In Series III, in which the task was drawing down to a point, 
the special instructions as to the rate of drawing were given. 
The subjects are differently affected by these instructions. A. is 
least affected, and as his pressure records are unsatisfactory very 
little can be drawn from his results, except that the introspections 
show that an attitude of certainty is always accompanied by a 
clear visual image. C. is most affected by the special instruc- 
tions, the instruction to draw fast resulting in a large increase, 
and the instruction to draw slowly, resulting in a large decrease 
in the number of attitudes of certainty reported. The pressure 
records throughout for this subject show a correlation between 
certainty and a definite type of curve. With F. no noticeable 
change in the number of attitudes of certainty is brought about 
by the instruction to draw at a slow rate, but with the instruction 
to draw at a fast rate the number of attitudes of certainty is in- 
creased. This subject draws at a slower rate naturally than any 
of the others as shown by his records throughout. With this 
subject no correlation was found between certainty and a definite 
type of pressure curve in the series in which there were no special 
instructions and in the series in which the instruction was to 
draw at natural speed; but in the series in which the special in- 
structions were to draw at a slow or at a fast rate, this correla- 
tion was found. With this subject in Series II and Ila it was 
found that large differences in subjective certainty were reflected 



238 JOHN TRUMBULL METCALF 

in the pressure records, but that for small differences no cor- 
responding objective differences could be determined. So also 
here, it seems that the instructions "slow" and "fast," increasing 
as they do the subjective differences in attitude, make the ob- 
jective changes in pressure great enough to be recognized. This 
subject drew with greater pressure than any of the others, and 
seemed throughout less sensitive than they to small differences 
in pressure. The results of this series with R. are quite analogous 
to those already obtained with him in the previous series. Cer- 
tainty is found to go with a long line, a fast rate of drawing, 
and a definite type of pressure curve, and as before with him 
the accuracy of drawing seems to have no effect upon his cer- 
tainty. The individual differences exhibited by the subjects in 
this series are interesting in that they seem to show that persons 
of different imaginal type are affected in different degrees by the 
same special instructions. 

Series IV gave an unexpected result. It was thought that 
when the right hand was added in the drawing process greater 
certainty in the drawing made with the left hand would be the 
result of the better motor control afforded by the symmetrical 
movement. It was found that the reverse was the case, the sub- 
ject reporting less certainty when the right hand drew with the 
left than when the left hand drew alone. The objective records 
show that there is little difference in accuracy, and no definite 
difference in the pressure characteristics between the two series 
of left-hand drawings. The time records, however, show that 
the drawings made by the left hand when the right hand is draw- 
ing symmetrically with it are made slightly more slowly than 
when the left hand draws alone, except in the case of C. Thus 
we find that the only difference in the objective records made by 
the change in the total form of the reaction which comes with 
the addition of the right hand is a slight increase in time in the 
case of three of the four subjects. Nevertheless, the introspec- 
tive reports betray that with this change in the total form of the 
reaction there comes a change in attitude in the direction of less 
certainty. This is described subjectively as a shifting of the 
attention from one drawing, or one part of the reaction, to the 



EXPERIMENTAL STUDY OF CONSCIOUS ATTITUDES 239 

other. Thus, with a change in the total reaction we find a change 
in attitude. Subject C, who was least affected as to his attitude 
by the change, is also the only subject whose time is not length- 
ened, and he is the subject who throughout the investigation has 
been found to use chiefly kinaesthetic imagery. Here again, 
therefore, we find an instance of imaginal type characterizing 
the degree of change of attitude under special conditions. 

The general result of our investigation is positive. In spite of 
numerous exceptions and some incomplete records, the results 
are sufficiently uniform and definite to show that under the con- 
ditions of our experiment some objective characteristics of the 
subject's reaction vary with his introspective attitude. The ac- 
curacy of the drawing seems less important than the other char- 
acteristics for the appearance of certainty. In some cases greater 
accuracy and certainty go together, but more often with the same 
subjects they do not. As for the time factor, certainty is usually 
found to go with a faster, and especially with an accelerating rate 
of drawing. Certainty is found also in most cases to be corre- 
lated with a definite type of pressure curve, and uncertainty 
with a variation from this type. In some cases, to be sure, notably 
with subject F., this correlation does not hold between small dif- 
ferences in attitude, but since it holds between large differences, 
this lack of complete correlation may be attributed to the crude- 
ness of the apparatus which is not sensitive enough to show 
slight objective changes. In Series IV and IVa we find no ap- 
preciable change in the pressure curves of the left-hand drawings 
— the form of the left hand's reaction is the same in both cases. 
In this series the difference in subjective reaction is, however, 
correlated with the difference in the subject's total reaction which 
is brought about by the introduction of his right hand into the 
process. 

The objective characteristics that we have measured with our 
apparatus, though they are directly involved in carrying out the 
instructions, are by no means the only ones that might vary with 
the changes in subjective attitude. Unconscious eye-movements, 
incipient movements of the limbs (especially of the arm and 
hand not used in the drawing), innervations of the larynx, etc., 



240 JOHN TRUMBULL METCALF 

might all be part of the subject's reaction to the instructions and 
might vary with his attitude, just as we have been able to show 
that certain characteristics of the movements most obviously 
involved in his reaction vary with changes in attitude. Since 
it is possible that any of these other motor processes might have 
been as closely and perhaps more significantly correlated with 
the attitudes of certainty and uncertainty, the writer hesitates 
to claim anything final for his results. The investigation is re- 
garded rather as a tentative, exploratory one. The results in 
his estimation do demonstrate that the conscious attitude is ac- 
companied by relevant objective differences in bodily reaction, 
and he ventures to express the hope that other investigators will 
be encouraged to follow up a line of investigation which appeals 
to him a promising one. 

REFERENCES 

i. 1 Freeman, F. N. Preliminary experiments on writing reac- 
tions. Psychol. Rev. Monog. Suppl., 1907, 8, 301-309. 

2. Judd, C. H. Movement and consciousness. Psychol. Rev. 
Monog. Suppl., 1905, 7, 199-226. 



COMPLEX REACTIONS OF THE DOG: A PRELIMI- 
NARY STUDY 1 

Arthur Howard Sutherland, Ph.D. 

Instructor in Psychology 

Yale University 

This study, originally begun with the hope of approaching the number 
processes (counting, estimating, or reacting to number) began with an at- 
tempt to set up rhythms of action on the part of the dogs. Its scope as here 
reported covers the preliminary stages in the process of learning, and be- 
came, necessarily, a study of methods. The apparatus used can be manipu- 
lated from outside the experimental room, and is fully described. The 
training series is described in a way which has not heretofore been done, 
and is advocated as an equalizer of animals which inevitably differ in a 
large number of respects. Certain observations regarding the transfer of 
habits learned in one series, and of the retention, seem important enough 
to present at this time. The method of treating the data of the experiment 
has also some novel features and leads to the conclusion that the evidence 
for discrimination is to be looked for in the integration of acts, which 
demands a special display of results, rather than a table of summary aver- 
ages of accuracy. 

Two pertinent criticisms have come from the Chicago labora- 
tory with regard to current studies of discrimination in animals. 
One deals with the method of treatment of data, the other with 
the nature of the stimuli. Hicks and Carr (3) have shown that 
the percentage method of tabulation of results of learning does 
not accurately nor adequately represent the activities of the 
animals under investigation. It follows that a percentage of 
error curve is, for psychological purposes, an uncertain datum 
for inferences. And Miss Weidensall (10) has shown that when 
a black and a white surface are presented to an animal (for dis- 
crimination), the white surface is of greater importance in 
modifying the behavior than is the black, and that what appears 
to be a discrimination may prove to be a "simple recognition." 
If these two criticisms are accepted, it is clear that the evidence 

1 1 wish to acknowledge with pleasure my indebtedness to Dr. J. M. Flint 
of the Department of Surgery in the Yale School of Medicine, for the use 
of dogs and quarters in which to carry on this experiment. 



242 ARTHUR HOWARD SUTHERLAND 

for discrimination in animals is seriously undermined; also that 
the treatment of the results of behavior must be revised if dis- 
crimination, in the psychological sense, is to be 'demonstrated. 

With a view to contributing to this question, the present pre- 
liminary study was undertaken as an approach to a fuller investi- 
gation of more complex processes of intelligence, especially the 
so-called "number" or "counting" processes. The experiment 
has failed to reach a point at which any contribution can be made 
to the latter problem, and this report presents only the behavior 
in the beginning stages of learning. 

It is possible to exhibit behavior of animals in different ways. 
If the experimenter is interested in a statistical problem, per- 
centages of error, averages, means, standard deviations, etc., 
are important, and from them certain conclusions can be drawn 
regarding the adaptation of animals to a given situation. But 
the successes and failures, the right and wrong choices, sum- 
marized in a percentage, conceal rather than exhibit the actual 
behavior of an animal in response to a controlled stimulus, and 
encourage generalized statements regarding groups of animals. 
In analyzing specific varieties of behavior, therefore, such as 
that involved in discrimination, this concealed behavior must be 
taken account of. And since the effort is to discover whether 
a certain type of internal behavior accompanies the external 
manifestations of choice, it would appear that a mere "Success 
or Failure" summary of external manifestations must inevitably 
fail to yield the desired data. Conclusions as to animal dis- 
crimination have therefore been mainly negative and, when posi- 
tive conclusions were drawn, reviewers have frequently sug- 
gested a faulty technique. 

But if an adequate description of the actual behavior were 
given, certain positive data would be available as to the processes 
of the animals. Hence, more complete descriptions rather than 
the experimenter's summary conclusions are needed in order that 
the science of behavior may progress by cumulative results. Not 
"do animals or these animals discriminate?" but "what processes 
are now going on in this animal ?" is the pertinent question. And 
only complete descriptions, in other than percentage terms, will 



COMPLEX REACTIONS OF THE DOG 243 

yield the desired data. The present report, therefore, is con- 
cerned with the behavior, and the analysis of behavior, of dogs in 
learning a visual discrimination. The rate of progress of the 
learning suggests a slow integration of various acts of the be- 
havior, and the presentation emphasizes this by showing the rela- 
tion of the choices to an integrated rhythmic activity. 

Historical 

Johnson (6) has reviewed critically the experimental work of 
Kalischer, Munk, Rothmann, Swift, and others. In his own 
dogs, he was able to discover no conclusive evidence of discrimi- 
nation of pitches, but succeeded in setting up an association be- 
tween food and noises, in which the localization of the noises 
apparently controlled the reactions. E. M. Smith (7) found his 
dogs able to make color discriminations "weak, unstable and 
easily inhibited by differences of luminosity." Colvin and Bur- 
ford ( 1 ) found that changes of form inhibit color discrimination 
in dogs. Hunter (5) succeeded in setting up a visual discrimi- 
nation (?) in the case of two dogs to the extent of J 2% and 
60% accuracy out of 560 and 650 trials. The experiment was 
characterized by "helplessness on the part of the dogs" and a 
"lack of resourcefulness." The position of the dog's head at 
the moment of release seemed to determine the direction of move- 
ment. Successive choices tended to encourage the maintenance 
of an orientation motor attitude for periods of increasing length. 
Franken (2) had earlier characterized similar reactions as "sen- 
sory thought processes." The use of this rubric is justified by 
Franken from the fact that the reaction is the release of an ex- 
pectant attitude — as contrasted with reactions which are im- 
mediate (p. 51). If the animal possesses expectant attention, it 
possesses the necessary rudimentary organization of intelligence 
for thought processes. 

The Dogs 

Six dogs, three male and three female, were used. The animals 
are referred to as BTF (Black Terrier, female) ; BF (Black, 
female) ; BrF (Brown, female) ; BM (Black, male) ; BBM 
(Brown Bull, male) ; WM (White, male). These dogs were 



244 ARTHUR HOWARD SUTHERLAND 

mongrels and had been confined to the kennels for varying 
periods. The precise ages and experience of the dogs were un- 
known. In the kennel the males were fighters ; the females were 
smaller, younger and more docile. 

The use of dogs picked up from the street has been criticized 
by many authors. The chief objection seems to be that on ac- 
count of a lack of knowledge regarding the previous experience 
of the animals, it may be that the behavior observed in the ex- 
periment is simply a repetition of behavior already learned else- 
where, and that the experimental conditions do not control the 
action, and particularly the learning. But the same objection 
may be urged against two dogs which have been reared in the 
laboratory. There is no typical dog. Each one shows marked 
individual differences. It seems desirable, therefore, to devise 
a method which shall equalize these differences to some degree, 
and bring the animals to a common basis for comparison. Not 
only is this necessary in order that the grouping of results from 
a single experiment may be valid; but it is the only means by 
which various pieces of work may be adequately compared. So 
far as is known this equalizing has not heretofore been done 
systematically. Vague references to a training series leave one 
in the dark as to the steps employed. In the present experiment 
a simple alternating habit was developed as a basis of compari- 
son, and the growth of this rhythmical habit is fully described. 
And while this method is believed to equalize the dogs upon a 
level for a psychological experiment, we have preferred to hold 
these results as tentative until the method can be used upon 
young dogs. 

Notes on Behavior in Learning the Alternating Habit 

In view of criticisms relative to the presence of an experi- 
menter or master in the vicinity of the dog while choosing, and 
to rule out the possibility of unconscious signs from which the 
dogs might take the cue, it was so arranged that from the begin- 
ning of experimental work proper (after preliminary training), 
the experimenter was absent from the room in which the reactions 
and choices took place. This necessitated a training series which 



COMPLEX REACTIONS OF THE DOG 



245 



was carefully watched. For this purpose Apparatus A, a modi- 
fled form of the Yerkes light discrimination box, was used 
( Fig. 1 ) . The apparatus was reduced to two parallel alleys open 
at one end. During the first month the animals were trained by 
regularly progressive steps to pass alternately from one alley to 
the other, securing in each a piece of meat. 



I* 



[^ 




r 



/ 



Fig. 1. — Apparatus A 



Nov. 20, 1 91 4. Each dog was taken to the apparatus room 
in turn, and fed at the food box in each alley of the apparatus. 
All the animals betrayed signs of awareness of the presence of 
food, three coming again and again to the vicinity of the meat, 
standing rigidly and with lips wrinkling. Each dog was per- 
mitted to investigate the room and apparatus so far as he would. 



246 ARTHUR HOWARD SUTHERLAND 

When the lights were extinguished, the dogs crouched close to 
the experimenter's feet. All tended to follow the experimenter's 
movments, but in different degree of proximity. BF, BM, and 
WM gave evidences of fright. It was found necessary to drag 
them into the apparatus and up to the food box, and immediately 
after snatching the meat, they would back away to the room out- 
side the apparatus. 

Nov. 21. Same. Less appearance of fright. 

Nov. 22. WM still showed signs of fear and had to be dragged 
to the food box. All others stood quietly by the food box. 

Nov. 23. With a piece of meat held just before the dog's 
nose, the dog was led to and fro about the room, and finally into 
the apparatus, being fed only at the food box. WM, this day, 
seemed less fearful, but it was still necessary to pull him to the 
apparatus occasionally. 

Nov. 24. Same. BrF, introduced to the experiment at this 
point (to take the place of another dog which had died), showed 
no signs of fear. The experimenter took his stand in the center 
of the apparatus and led the dogs from alley to alley. 

Nov. 25. Same. 

Nov. 26. Same. Dogs now ran around behind the experi- 
menter from one alley into the opposite alley without much 
coaxing. 

Nov. 27. Same. Dogs would start to run to the opposite side 
if anything was dropped on the floor. They apparently waited 
for a cue. 

Nov. 28 to Dec. 3. Same. If meat was not in evidence when 
the dog arrived, he would scratch at the trap door. 

Dec. 4. Food loaded at opposite ends of alternate shelves of 
the chute (Fig. 2), so as to drop first in one alley, then in the 
opposite. The experimenter took his stand outside the apparatus 
without disturbing the reactions. 

Dec. 5. Same. WM gave no evidence of fear. 

Dec. 6-10. Same. Dogs proceeded rhythmically from one 
alley to the other. 

Dec. 11. Experimenter took his stand at N, the doorway of 
the room. BrF, WM, BF, BTF each stopped several times to 



COMPLEX REACTIONS OF THE DOG 247 

look at the experimenter, but proceeded without overt en- 
couragement. 

Dec. 12-16. Same. 

Dec. 17. During this day's run the experimenter backed still 
further out of the room without inhibiting the dogs' behavior. 

Dec. 18-20. The experimenter did not advance into the room 
during these three days. During parts of the run, he stepped 
out of sight, being guided in the operation of the apparatus by 
the sounds of the dogs' movements. 

It will appear from these notes that one month was consumed 
in setting up an alternating habit which would run its course 
during each day regardless of the presence of the experimenter. 
Whatever cues may have set off the adjustment, they did not 
come from the experimenter. In bright light, as also in a dim 
light which just permitted the dog to be seen, a pendular-like 
swing was set up, with a short stop at each end of the swing, 
and continued until no more food was found. 

No systematic tests were made as to the length of time the 
rhythm persisted after the failure to secure food. Since the 
swings did continue, however, the cue to the continuation of the 
rhythm was not invariably the just prior "act of securing the 
food." Sight, sound and odor of food about to be secured like- 
wise were excluded. But sights and odors, so far as they referred 
to the total situation, were not excluded, and it is quite possible 
that these, together with the kinaesthetic and organic qualities 
involved in approaching the position at which the food is usually 
obtained, are to be looked to for explanation of this behavior. 
To make a general test of the effect of the presence of a larger 
amount of odor, the front of the chute was removed so that all 
the meat except that on the four lower shelves was exposed. No 
effect could be observed in the dogs' behavior. 

The experiment with controlled stimuli begins, then, at this 
point (with an unknown number of influential factors) and 
should proceed analytically in the control and variation of each. 
This is of course similar to the case of any human or other animal 
experiment. In the human being, a complex set of fairly well 
automatized language habits takes the place of this simple alter- 



248 



ARTHUR HOWARD SUTHERLAND 




COMPLEX REACTIONS OF THE DOG 249 

nating habit, and by the variations of the verbal expressions the 
experimenter infers the experience of his human subject. The 
alternating habit of the dogs is far less complex but has the same 
function, so that in the present experiment we have the advan- 
tage of a single simple adjustment, the variations of which will 
betrav the relative influence of the stimuli used. It now becomes 
necessary to determine whether this simple alternating habit can 
be broken up or complicated by means of such stimuli, without 
the presence of the experimenter. 

Herewith (Fig. 2) is shown a perspective drawing of the ex- 
perimental room and the apparatus. The walls of the room were 
lined with black cloth, and the starting box (C) and the doorway 
were also covered so as to exclude any visual stimuli from the 
outside. The dog was brought to the starting box, and there 
confined until the apparatus was adjusted. The doors, D and 
D', were then pulled back, the small door, N, releasing the animal 
into the vestibule through a narrow passageway which necessarily 
gave him a direct orientation toward the two alleys. Having 
passed into one of the alleys, the doors D and D' were closed 
behind him, the further pathway leading through the exit, E or 
E', through the return alleys, A' or B', through the small doors, 
c or c', back to the starting box, C. 

After a few days' experience, it was found no longer necessary 
to close the doors behind the dogs, since they proceeded along 
the pathway back to the starting box. The reactions therefore 
involved a more or less continuous run (for sixteen choices), the 
animals stopping only to secure the piece of meat at the foot 
of the chute. 

In greater detail the additions to the apparatus as found in Fig. 2 may be 
described under the following headings, Food Chute, Stimulus Lights and 
Wiring, Reflectors, Screen, Dumping Apparatus, Movable Partitions, Start- 
ing Box, Operation of Doors, Curtains. 

Food Chute: In order to drop a piece of meat at the appropriate spot in 
the apparatus, a chute was devised, CH. This is an enclosed box 5 ft. x 6 in. 
x 6 in. set upright midway between the two alleys, in an opening at the end 
of the mid partition. Within the box are two horizontal shafts, over which 
runs a broad endless belt. The belt carries 16 shelves, each 1 in. deep and 
5 in. long, set so that each end hangs over the corresponding alley. The 
shelves are 3 in. apart upon the belt. Attached to the outside end of the 



250 ARTHUR HOWARD SUTHERLAND 

upper shaft is a pulley and over this pulley runs a cord to another pulley 
outside the experimental room. When the cord is pulled, the belt carries 
the shelves downward and as each one reaches and passes around the lower 
shaft, it deposits its contents at the foot of the chute. Before the dog was 
brought to the experimental room, the chute was loaded, by placing sixteen 
pieces of meat (rolled in cornstarch so that they would slip) upon the 
shelves in such a way as to drop at F and at F' in an order determined on 
for that day, or series. The experimenter may thus stand outside the 
experimental room and while the dog is returning to the starting-box, drop 
another piece of meat in the same or the opposite alley and be ready for 
the succeeding reaction. 

Reflectors. Two mirrors, at M and M', were- set to reflect an image of F 
and F' to a hole in the wall at the experimenter's station. From this station 
the experimenter therefore could observe any failure of the apparatus. To 
intensify the brightness of the image of F and F', small mirrors were set 
at 'R and R', at the top of the side partitions of the alleys A and B. These 
reflect a beam of light from the stimulus lights to the foot of the Food Chute. 

The Stimulus Lights. A row of five sockets for no v. finger size heat 
lamps was placed upon a support, and this support attached to the wall at 
the end of the alley. The lamps were wired in parallel so that one or five 
lamps could be used. The tops of the lights were visible through a rectangu- 
lar opening, six by four inches. The number of lights used differs for the 
several series and is discussed in that connection. In the later series, one 
or more lights were shown in both alleys, and to facilitate the manipulation 
of the stimulus lights, a special" wire was run directly from the main current 
to the number of sockets necessary. A double knife switch upon the wall 
outside of the room at the experimenter's station, controlled the remainder 
of the lights, so that a single throw of the switch turned out the lights in 
one alley, except those on the direct wire, and turned on the corresponding 
lights in the other alley. It is the stimulus lights, L and L', which reflect 
from the small mirrors, R and R', to the foot of the Food Chute. This 
serves the double purpose of giving to the experimenter a brighter image of 
the food position, and also enables the dog more easily to locate the piece 
:>f meat. 

Screen. A small wooden screen at S and |S'' likewise serves a double 
purpose of shielding the food from the dog until he has entered the alley 
far enough that the doors may be closed behind him, and also as a scrape 
in connection with the dumping apparatus. 

The Dumping Apparatus. If the dog chooses the wrong alley, i.e., the one 
from which proper stimulus and food are lacking, the piece of meat must 
be removed before the next reaction occurs. To accomplish this conven- 
iently, a hole through the floor at the foot of the chute, is covered with a 
paddle shaped trap door, IF and F\ This trap door slides under the screen, 
S and S', and its contents are scraped off and fall through the hole in the 
floor. The mechanism for operating the trap is a long arm, f and t\ which 
slides along the floor between guides. At g and g' an upright arm rests in 
a malleable iron socket upon the long arm. The upright arm is fixed at its 
center to the mid partition, its upper end being free. To operate the trap, 



COMPLEX REACTIONS OF THE DOG 251 

therefore, it is only necessary to push forward upon this upright lever. 
Reversing the movement closes the trap again. 

Movable Parts. To give the animal the best possible orientation toward 
the lights, as it leaves the starting box, the right angle parts, P and P', were 
devised. These are pushed back out of the way when the apparatus is not 
in use. The doors, N, c and c', are attached to these movable parts. 

The Starting Box, C, is a three sided, hinged box, the open side being 
directed toward the apparatus. The top is covered over with cloth so that 
an animal within cannot see what is going on around the box. This covering 
also extends up over the doorway of the room during the course of the ex- 
periment. To open and close the doors, c and c', a slender rod 2 was fastened 
to the top of each, and passed through holes in the wall of the room, to the 
experimenter's stand. The same method is used for working the levers 
of the dumping apparatus. The doors, D and D', are kept closed by the 
springs, Sp and Sp', and are opened by pulling upon a cord which branches 
at its mid point, sending one branch to each door. 

No means of punishment has been used in connection with the 
experiment thus far. For purposes of this preliminary series, 
absence of punishment was decided on in order to introduce no 
factors which might prove to be inhibitory of the dog's reactions. 

First Series 

In the first series, one light was shown, at L or LA Meat was 
to be found at F or F' — in the alley in which the light was shown. 
The overhead light was turned off so that the room was dark 
except for the single stimulus light. The light was shown, for 
successive choices, in the following order (B = Right alley, 
A = Left alley) : B, B, A, A, B, A, A, B, A, B, B, A, B, B, 
A, A — 16 choices, half in alley B, and half in alley A. Will the 
alternating habit of the preliminary series persist, or will the dog 
show a preference for the light or for the dark alley? The only 
respect in which the conditions were changed from the preceding 
training series was in the location and intensity of light. 

During the first ten days of this experiment, the dog was held 
at N by a long leash extending through and hooked to the wall 
(instead of in a starting box, as later). When the apparatus 
was ready for operation, the loop in the leash was pushed from 
the hook, the doors, D and D', opened so that the light from the 

2 A part of the apparatus devised by the writer in connection with some 
work on Color Vision in Cats in 1907 with the Watson Spectral Light Ap- 
paratus, — unpublished. 



252 ARTHUR HOWARD SUTHERLAND 

stimulus lamp shone upon the dog, which then leaped forward 
into the apparatus, either alley B or alley A. By means of the 
long leash, the dog was then returned to N for another choice. 
After the tenth day, the dog followed on his own account the 
pathway through the return alleys to the starting box, C, from 
which it was released by opening the door, N, further to guard 
against possible influence of the dog by the experimenter. 

For purposes of comparison, the percentage of error curves 
are given (Fig. 3) for each dog, and an average curve for the 
group of 6 dogs. They show that on the first day all of the 
dogs except one (WM) were successful in more than 50% of 
the choices, varying between 56.25% and 68.75%, WM being 
successful in 37.25% of the choices. On the second day, two 
of the dogs were less successful, one equally, and three more 
successful than on the first day. On the third day, all except 
one were more successful than on the second day. On the fourth 
day, one dog ran a perfect series, four improved, one held his 
record of the day before. On the fifth day, two were more suc- 
cessful, four were less successful, etc. 

These records show that the discrimination of a light from a 
dark alley, widely different as these seem to be, is a difficult 
undertaking under the present conditions. Is this difficulty due 
to the previously established alternating habit? The percentages 
conceal the, evidence on this question, and another means must 
be had to exhibit the actual behavior of the dogs. Behaviorism 
has not yet devised a terminology and the difficulties are there- 
fore enhanced. 

The day's run in the present series consisted of 16 choices; 
and after the first choice, the ensuing behavior may be brought 
under two lesser types — respectively "succession" and "alterna- 
tion." .Having made a choice of alley A (let us say), if the dog 
repeats his trip to alley A after returning to the starting box, he 
is credited with a "succession." If, however, the dog goes on 
the second choice to alley B, he is credited with an "alternation." 
The behavior in each series may thus be stated in terms of "suc- 
cessions" and "alternations." 

The lights were shown in an order which, if perfectly followed, 
would have required 6 "successions" and 9 "alternations." In 



COMPLEX REACTIONS OF THE DOG 253 

the preliminary training series, the dog had learned to make 15 
* 'alternations' ' per day. On the first day of the present series 
the "succession" type of behavior predominated, as follows : 









Table I 














No. 




% 




Sum 


Dog 


Behavior 


Success 


ful 


Successful 


%of 15 


total % 


BTF 


Sue. 


11 


7 




.636 


.46> 






Alt. 


4 


3 




•75 


.20 


.667 


BF 


'Sue. 


12 


7 




.58 


.467 






Alt. 


3 


3 




1. 00 


.2 


.667 


BrF 


Sue. 


9 


5 




•55 


.33 






Alt. 


6 


3 




•50 


.2 


•53 


BBM 


Sue. 


5 


2 




.40 


13 






Alt. 


10 


7 




.70 


.467 


.60 


BM 


Sue. 


9 


5 




•55 


■33 






Alt. 


6 


3 




•50 


.2 


•53 


WM 


Sue. 


8 


3 




■37 


.2 






Alt. 


7 


3 




•43 


.2 


.40 



<( 

(( 



It is at once clear by reference to the above table, that the dif- 
ficulty does not lie in the persistence of the alternating habit, 
except in the case of BBM. 

Transfer of Training 

The first experimental series, from the human point of view, 
required merely a slight modification of the preliminary series. 
It was different from the preliminary training series by the num- 
ber of "successions" required, in place of a similar number of 
alternations." The "successions" were interpolated among the 
alternations" in the following order — S, A, S, A, A, S, A, A, 
A, S, A, A, S, A, S. The change of location and intensity of 
light made the situation sufficiently different to inhibit the trans- 
fer of the alternating habit. Was there, however, a position 
error on the first day of the series, and if so, on which side? 
My records show, 1st, that any position error present the first 
day was eradicated after the third day (BTF made a perfect score 
on the fourth day) ; 2nd, that BBM, whose tendency was to con- 
tinue the rhythmical habit on the first day, showed more "suc- 
cessions" than "alternations" on the second day; 3rd, that BF 
and BM developed a pronounced tendency toward "successions" 
on the third day. A further examination of the records shows 
that this last apparent tendency was a sporadic affair which per- 
sisted only for the one day. 



254 



ARTHUR HOWARD SUTHERLAND 







2 
pa 



l 



1 CO 




o 






1— • 






CO 






►"• 






f^ 






H-f 






vO 






«•— 






m 




u 






O 


i ^j- 




u 


- 




<y 


c*l 




<-M 




> 

»— ( 


>> 


>— * 


> 
1* 


II II 


O 


a 


ites 
sac 


<^ 




a -a 


oo 




»0 CO 
S-c rQ 


IS* 




o< 


sO 






m 






^r 






rn 






M 







£2 P *o ,53 co iriH k^o 

^" tn vc^O vo rsco CO O — 



^ m jn. w »jT> i->. q 

AVO fM 00 in « NfOO 
in ur>\0 VD r^OOOO O .— 




4J 



~ u 



CQ 



lo O in in O in 
oi m h^ ex. "^ r> c 

6 \p oi co* in •-" r> £? O 

in utjVO O r-sCC CO C\ •- 




<1> 

> 

O 



in o in mom 

« ION . °i ^ *> O 

o'O oico m «-« Nf^O 

iO in so so t^CO CO On ~ 



COMPLEX REACTIONS OF THE DOG 



255 














►*' ^ 



2 

u 




•o 



> 

> 

U 

O 



O 

1—1 



^ 












O in ih^O ^O NCO CO O.-^ 1 - 1 



to rf 10 ir.vo \C J^CO OD O ~ 



256 



ARTHUR HOWARD SUTHERLAND 









Table II 




Successions 






on L 


R 


First Day 


BTF 


ii 


o 




BrF 


4 


5 




BF 


ii 


i 




BBM 


3 


2 




BM 


o 


9 




WM 


7 


2 


Second Day 


BTF 


12 


1 




BrF 


7 


2 




BF 


8 


2 




BBM 


2 


6 


- 


BM 


3 


4 




WM 


8 


i 


Third Day 


BTF 


9 


o 




BrF 


3 


4 




BF 


13 







BBM 


2 


6 




BM 


14 


o 




WM 


7 








Alterations 
L to R R to L 



2 

3 

i 



3 

2 
I 

3 

3 
3 
4 
3 

3 

4 
i 

3 

i 

4 



2 

3 

2 

5 
3 
4 

I 
3 

2 

4 

4 
3 

3 
4 
I 

4 
o 

4 



Learning to Discriminate 

In learning a series, the behavior is modified, not by success 
in general, but by each particular success or failure. The evi- 
dence for discrimination consists in the integration of a series 
of acts in a definite order (when the order of presentation of 
stimuli is constant) and in the variation of this order (when the 
order of presentation is varied). The problem here is not so 
much to determine whether the animal does discriminate, as to 
determine the particular conditions under which this modification 
of behavior occurs, to note how rapidly the dogs respond to the 
success or failure to obtain reward, and to note the type of be- 
havior which follows each particular success or failure. The 
following table shows the complexity of the progress, for the 
dog BTF. In the first group of columns are shown the num- 
bers of cases in which a successful choice on the left led to a 
successful "succession," and to a successful "alternation" to the 
right; also the choices successful on the right which were fol- 
lowed by a successful choice by "succession" or "alternation." 
(S = "succession"; lAr = "alternation" from left to right; 
rAl = "alternation" from right to left.) The second group 
shows the choices which were successful by means of "sue- 



COMPLEX REACTIONS OF THE DOG 



257 



cessions" and "alternations," when the preceding choice was un- 
successful. The third group shows the unsuccessful behavior 
which followed a successful choice. The fourth group shows the 
figures for unsuccessful behavior when the preceding choice was 
unsuccessful. 

Table III 
Dog BTF 



Day 


Successful 


Unsuccessful 


Successful 


Unsuccessful 


of 


followed by 


followed by 


followed by 


followed by 


Series 


Successful 


Successful 


Unsuccessful 


Unsuccessful 




L 


R 


iL 


iR 


L 


R 


iL 


<R 




S lAr 


S rAl 


S lAr 


S rAl 


S lAr 


S rAl 


S lAr 


S rAl 


1 


3 2 


1 


4 





2 


1 


2 





2 


3 1 


1 1 


4 





3 





2 





3 


3 3 


1 


4 





1 


2 


1 





4 


3 4 


3 5 




















5 


3 1 


1 4 


1 2 





3 











6 


2 3 


1 2 


3 





1 


2 





1 


7 


1 3 


2 4 


1 


1 


1 1 


1 








8 


3 2 


2 4 


1 1 





2 











9 


3 3 


2 3 


1 1 





2 











10 


2 4 


3 4 





1 





1 








11 


2 4 


3 4 





1 





1 








12 


3 3 


3 4 


1 





1 











13 


3 4 


3 5 




















14 


3 4 


3 5 




















15 


3 4 


3 5 




















16 


3 4 


3 5 




















17 


2 3 


3 5 





1 


1 











18 


3 4 


3 5 




















19 


3 4 


3 5 






















51 60 42 J2 


15 5 


1 3 


15 3 


2 6 


5 


1 



This table shows also the progress of integration. There is a 
gradual concentration of the figures under the "Successful to 
Successful" column, and a gradual diminution of the figures in 
the other columns, with sporadic lapses. The table shows in 
some detail that this progress is accomplished by elimination of 
a certain number of reactions under one type of behavior, the 
substitution of the other type of behavior for it, and at the same 
time the ordering of the precise sequences according to the estab- 
lished order of stimulus presentation. The behavior thereupon 
must be considered, not merely behavior, but behavior with re- 
gard to a discriminated object, or sign. 

Similar tables have been prepared for each dog, but as will 
appear from our summary of the problem and difficulties, there 



258 



ARTHUR HOWARD SUTHERLAND 



are good reasons for omitting them at this time. The purpose 
of including the above table is to indicate the method of pre- 
senting the results which will show the complexity of the process 
of integration of behavior which is necessary as a criterion of 
discrimination. By means of this table the elimination of un- 
successful "successions" and "alternations" and the points at 
which one type is substituted for another are shown. Taken in 
conjunction with the following Table IV, which shows the days 
on which each error occurred, an exhibit of the detailed behavior 
is given which retains the relation of each choice to the integrated 
behavior. This is valuable for psychological purposes. 

Was the integration of the behavior of equal difficulty through- 
out, or were there points in the series which required a longer 
time to master? The following table shows, for the same dog, 
BTF, that the first choice was mastered at the fourth day, and 
no errors were made thereafter. The same is true of the second 
choice. On the third choice an error occurred on the 7th day, 
etc. 

Table IV 



•Choice 


Dog 


Days on which errors occurred 


R 


L 



Total 


1 


BTF 


1, 2, 3 


3 


3 


2 




1. 2, 3 


3 





3 


3 




7 





1 


1 


4 




17 





1 . 


1 


5 




1, 2, 3, 5, 12 


5 





5 


6 




10, 15 





2 


2 


7 




9 





1 


1 


8 




2, 6, 8, 9, 13 


5 





5 


9 
















10 




1, 2, 5, 7, 8 


5 





5 


n 




1, 2, 3, 6, 9 








5 


12 




12 





1 


1 


13 




5 


I 





1 


14 




1, 3, 6 


3 





3 


15 




11 





1 


1 


16 




7 





1 


1 








30 


8 


38 



The table shows further the relative difficulty of the integra- 
tion of the various acts of choice. If a line were drawn con- 
necting the final one of the days on which errors occurred, it 
will roughly indicate the relative difficulty of each choice in the 
series for the dog BTF. It also shows that notwithstanding the 



COMPLEX REACTIONS OF THE DOG 



259 



alternating habit which preceded the series, a preference or posi- 
tion error persisted and is shown predominantly on the 5th, 8th, 
10th, nth choices in the series. Grouping by fours the 38 errors 
which this dog made in establishing this integration, it will be 
seen that the large proportion of the errors occurred in the 
middle of the series. 



Errors on the 
1st 4 
2nd 4 
3rd 4 
4th 4 



»R 


L 


Total 


6 


2 


8 


10 


3 


13 


10 


1 


11 


4 


2 


6 



38 



Another difficulty which is apparent is that of retention. For 
the first six days the third choice was properly integrated, on the 
seventh there was a slip at this choice. For sixteen days the 
appropriate type of behavior was called out on the fourth choice 
and during the nine days previous there had been no errors on 
the first four choices. On the seventeenth day, an error was made 
on this choice. Summing the errors for each choice, for the 
entire group of dogs we have the following Table V. (S = "suc- 
cession, A = "alternation.") 



Choice 









Table V 








I 


18 


errors 


Choice 


11 


29 


S 


2 


23 


S 




12 


7 


A 


3 


15 


A 




13 


16 


A 


4 


18 


S 




14 


19 


S 


5 


34 


A 




15 


11 


A 


6 


15 


A 




16 


16 


S 


7 


14 


S 






284 




8 


18 


A 










9 


5 


A 










10 


26 


A 











Of 284 errors made by the group, 136 occurred on the first and 
last four choices, 148 on the second and third four choices, bear- 
ing out the former statement that the middle of the series is more 
difficult. This may be taken as additional evidence that the 
alternating habit is not carried over, since in the first and last 
group of four, there are four "successions" and three "alterna- 
tions" and here a smaller number of errors was made than in the 
second and third groups of four in which there were two "sue- 



260 ARTHUR HOWARD SUTHERLAND 

cessions" and six "alternations." (These figures of course in- 
clude the 18 errors made upon the first choice.) "Successions" 
led to 119 errors, an average of 19 5-6 for each choice; and 
"alternations" led to 147 errors, an average of 16 1-3 for each 
choice. 

By way of summary of the results so far, it may be said that 
the apparatus described here admits of maniuplation from out- 
side a room; that dogs can be trained to work within the room 
without the presence of the experimenter; that a training series 
of the kind described is favored as an equalizer of various species 
of animals ; that this places the animals on a basis with reference 
to the experiment differing only in degree from that of the human 
subject whose language is behavior. As a result of the first 
series of this experiment it is shown that the alternating habit 
is not transferred to the situation which, so far as the experi- 
menter can control it, remains the same except for the position 
and intensity of the light; that a study of behavior demands 
behavior terms and the terms "succession" and "alternation" are 
suggested as descriptive of the behavior of the dogs in this ex- 
periment ; that a position error was developed but did not persist 
longer than three days; that the original behavior in this series 
was modified by the experience of the animals and the problem 
is to determine the conditions under which this modification oc- 
curs. Lack of retention is shown to be a source of difficulty. A 
series of 16 successive choices to be made according to an estab- 
lished order is shown to be of unequal difficulty in its various 
parts. Psychological study demands a mode of presentation of 
results which shall avoid vague general statistical summaries and 
get closer to the facts and a step in this direction is indicated by 
the tables. And finally it is proposed to define discrimination 
in behaviorist terms. The series of acts of choice which consti- 
tute the day's run may be thought of as a whole or as parts. The 
evidence for discrimination is to be looked for in the integration 
of the parts into a whole, with reference to some stimulus, which 
has become gradually organized as the representative of a par- 
ticular mode of behavior. 



COMPLEX REACTIONS OF THE DOG 261 

Test of Persistence 

At the close of Series 3, to be mentioned later, the apparatus 
was arranged for a repetition of the above conditions to test the 
persistence of the association set up. Twenty-two days had 
elapsed. All dogs followed the light, except BrF, who went once 
to the side on which no light was shown. On the second day, all 
dogs went to the light, except BM who went once to the dark 
alley. The third day all choices were correct. On the fourth 
day, the lights were shown in a new order : B, A, B, B, B, B, B, 
A, A, A, B, B, A, B, A, A. All dogs correctly followed this new 
order. We have here a new mode of behavior not the "suc- 
cessions" and "alternations" learned in a certain rigid order. The 
behavior has become flexible and is freed from the rigidity of the 
simple reflex. Something is selected and this selectivity is of a 
type which is the product of experience and fits only under the 
concept of discrimination. 

Second Series 

The second stage of the experiment required a similar dis- 
crimination between "two-lights-with-food" in one alley as 
against "one-light-without-food" in the opposite alley. The order 
of presentation of the two lights was the same as in the first 
series, viz., B, B, A, A, B, A, A, B, A, B, B, A, B, B, A, A, — 16 
choices. The series ran for nine days without appreciable change 
in the level of the curve showing percentage of error. BF be- 
came frightened at her image in the mirror during the first day 
and was dropped from the experiment. The results show that 
the dogs all started in with an excess of "successions" and that 
the general tendency was toward an excess of "alternations." 
The dog, BTF, on the first day made 10 "successions" and 5 
"alternations," of which 7 and 3 respectively were successful. On 
the ninth day she made 3 "successions" and 12 "alternations" 
of which 1 and 8 respectively were successful. Having been 
guided at this time by percentage of error curves which concealed 
the true integration which was slowly going on, I closed the series 
and began a third series. 



262 ARTHUR HOWARD SUTHERLAND 

Third Series 

In the third series "four-lights-with-food" as against "one- 
light-without-food" were used; and the order in which these 
were presented was twice in succession in the right alley, then 
twice in succession in the left alley. During the first four days, 
the dogs were permitted to run into the apparatus with the four 
lights showing and the other alley closed. This was designed 
to emphasize the four lights and to break up the tendency to 
"alternations" which had appeared in the second series. In the 
third series the curve of percentage of error remained level. 
There was constant fluctuation as to number and order of "suc- 
cessions" and alternations," but these varied from day to day 
and were not retained as permanent members of a growing com- 
plex. Such a fluctuation perhaps represents roughly the clawing 
and scrambling of animals in the problem box; or of running into 
blind alleys in the maze. But the retention of one type of be- 
havior and the elimination of unsuccessful behaviors failed to 
occur. There was a failure to pick up a cue which might be- 
come organized as the focal point of a complex of elementary 
acts. The series ran for ten days (160 choices for each dog). 
In the second and third series together (more than 300 choices) 
the difference in magnitude of light area (or in intensity of 
illumination, or in the number of lights) distinguished the two 
alleys. That this situation is very different for the dog, from 
the discrimination of a lighted from a dark alley, is at once clear. 
There is again a failure to transfer a habit from the first series 
to the following two. It is quite possible that a prolongation 
of the series might succeed in setting up the association, but the 
testing of the method was of greater concern at this time. 

Final Series 

In the final series, "four-lights-with-food" were shown in one 
alley as against "one-light-without-food" in the opposite alley. 
The order of presentation was the same from day to day, except 
that on one day the series began on the right, the following day 
on the left. The curve of percentage of error for WM is typical 
(Fig. 4). The series ran for 20 days (320 choices for each dog) 



COMPLEX REACTIONS OF THE DOG 



263 



and the accuracy varies between 50% and 81.25%. The accuracy 
varies little from that of the preceding series in which the first 
choice for each day was to be made on the same side. 

Although the accuracy of choice approaches that attained by 
the dogs of Hunter (5), the tables of analysis of behavior show 
clearly that no progress was being made and therefore there is 
no indication of discrimination. The following Table VI for the 
dog BBM shows the failure to progress in the distribution of 
the "successions" and "alternations" toward an expected stand- 
ard; and that while there was substitution of "successions" for 
"alternations" and vice versa, from day to day, there was no re- 
tention of the modifications. 

Table VI 















Dog BBM 


















Successful 






Unsucces 




Day 




Successful 


Unsuccessful 


;sful 


of 




to 






to 




to 


to 




Series 




Successful 




Unsuccessful 




Successful 


Unsuccessful 






L| R 






L| R 




L| R 


L| 


R 




S 


lAr| S rAl 


S 


lAr| S rAl 


S 


lAr| S rAl 


S lAr| S rAl 


1 








3 4 


7 





1 


203 





4 


004 








1 1 


2 





4 


1 


5 








044 








1 3 4 


2 





2 


3 


1 


3 


1 1 


6 


1 





124 





1 


3 4 


1 





1 


4 





3 





3 








044 








004 


4 





4 


5 





3 


I 1 


5 








1 3 4 





1 


3 4 


2 





2 


6 





3 





3 








044 








044 


4 





4 


7 


1 


1 


2 1 


5 


2 





305 





3 


025 











8 





2 





2 








4 4 


1 





3 4 


1 4 





5 


9 


2 


4 





6 


1 





023 








4 4 


2 





2 


10 





1 


1 2 


4 


1 


1 


2 1 5 





3 


025 


1 





1 


11 


2 


2 


1 2 


7 





1 


012 


1 


1 


1 3 


1 1 





1 3 


12 





2 





2 








1 3 4 








004 


4 


1 


O' 5 


13 





1 


3 2 


6 








3 1 4 





2 


013 


1 


1 


2 


14 





1 


2 3 


6 





1 


1 1 3 





3 


014 


1 





1 2 


15 





2 





2 





1 


124 





1 


1 3 5 


2 


1 


1 4 


16 





1 


2 2 


5 





1 


I 2 4 





2 


1 1 4 


1 





1 2 


17 


2 


3 


2 3 


10 





1 


012 





1 


012 








1 ] 


18 





1 


1 1 


3 








224 





1 


3 4 


3 


1 


4 


19 





2 


2 2 


6 





1 


1 2 4 





2 


1 1 4 








1 1 


20 





2 


2 2 


6 








1 2 3 





2 


024 


2 





2 


I 


8 


41 I 


25 26 




5 


8 


20 41 


2 


27 | 


4 46 


2 35 1 


4 


/ 



The total percentage of errors for the final series is approxi- 
mately the same as for the second and third. It is considerably 
more than 50%. This was an unexpected result since the change 
in the location of the beginning choice for each day would sup- 
posedly prevent the record from showing so high a percentage. 



64 



2h 



ARTHUR HOWARD SUTHERLAND 



W. M 




i 2 3 4 5 6 7«S 9 10 ii 12 13 14 >5 16 17 18 19 



Curve XXTII 
Figure 4 



The only conclusion which can be drawn regarding discrimination 
here is that, for the conditions of this experiment, the difference 
between four lights and one light is too small a difference upon 
which to organize and control the reactions. 

Attention should be called however to the fact that the re- 
actions occurred — perhaps in response to the total situation or 
to some part of the situation the nature of which is unknown — 
in such a way as to force the recognition of the underlying or 
ganic activity of the animals. The raw material, so to speak, U 
present; the influence of the preliminary training series is no 
shown in the specific mode of behavior, yet is shown in the cot 
tinuation of activity in a dark experimental room without th. 
. presence of the experimenter. Something has been carried over 
from the preliminary training series to the later experimental 
series which is not to be seen in the tables of results, but might 
be described as a "general set" toward experimental situation. 
Considering the difficulties of manufacturers to secure continued 
work from human beings unwatched, especially in new countries, 
it is worthy of note that the dogs will readily assume a "work" 
habit which leads them through a complicated apparatus day 
after day without immediate supervision. 

It is hoped that this experiment may be repeated under condi- 
tions more satisfactory for final conclusions.